cafkafk/lix
cafkafk/lix
1
0
Fork 0
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity Actions
lix/src/libexpr/parser.y

499 lines
14 KiB
Text
Raw Normal View History

* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%glr-parser
%pure-parser
%locations
%error-verbose
* Hack for Bison 2.3 compatability.
2006-10-02 16:43:15 +02:00
%defines
* Removed URIs from the evaluator (NIX-66). They are now just another kind of notation for strings.
2006-10-11 23:59:33 +02:00
/* %no-lines */
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%parse-param { yyscan_t scanner }
* Use a proper namespace. * Optimise header file usage a bit. * Compile the parser as C++.
2006-09-04 23:06:23 +02:00
%parse-param { ParseData * data }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%lex-param { yyscan_t scanner }
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%{
* Hack for Bison 2.3 compatability.
2006-10-02 16:43:15 +02:00
/* Newer versions of Bison copy the declarations below to
parser-tab.hh, which sucks bigtime since lexer.l doesn't want that
stuff. So allow it to be excluded. */
#ifndef BISON_HEADER_HACK
#define BISON_HEADER_HACK
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
#include "util.hh"
* Don't use ATmake / ATmatch anymore, nor the ATMatcher class. Instead we generate data bindings (build and match functions) for the constructors specified in `constructors.def'. In particular this removes the conversions between AFuns and strings, and Nix expression evaluation now seems 3 to 4 times faster.
2004-10-27 00:54:26 +02:00
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
#include "nixexpr.hh"
* Remove unnecessary inclusions of aterm2.h.
2006-09-05 00:08:40 +02:00
#include "parser-tab.hh"
#include "lexer-tab.hh"
* Make Nix build with Bison 2.4.
2009-01-12 13:51:28 +01:00
#define YYSTYPE YYSTYPE // workaround a bug in Bison 2.4
* Remove unnecessary inclusions of aterm2.h.
2006-09-05 00:08:40 +02:00
* Make sure that config.h is included before the system headers, because it defines _FILE_OFFSET_BITS. Without this, on OpenSolaris the system headers define it to be 32, and then the 32-bit stat() ends up being called with a 64-bit "struct stat", or vice versa. This also ensures that we get 64-bit file sizes everywhere. * Remove the redundant call to stat() in parseExprFromFile(). The file cannot be a symlink because that's the exit condition of the loop before.
2010-10-04 19:55:38 +02:00
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
* Use a proper namespace. * Optimise header file usage a bit. * Compile the parser as C++.
2006-09-04 23:06:23 +02:00
using namespace nix;
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
* Use a proper namespace. * Optimise header file usage a bit. * Compile the parser as C++.
2006-09-04 23:06:23 +02:00
namespace nix {
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
* Use a proper namespace. * Optimise header file usage a bit. * Compile the parser as C++.
2006-09-04 23:06:23 +02:00
struct ParseData
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
{
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
SymbolTable & symbols;
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
Expr * result;
* Use a proper namespace. * Optimise header file usage a bit. * Compile the parser as C++.
2006-09-04 23:06:23 +02:00
Path basePath;
Path path;
string error;
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
Symbol sLetBody;
ParseData(SymbolTable & symbols)
: symbols(symbols)
, sLetBody(symbols.create("<let-body>"))
{ };
* Use a proper namespace. * Optimise header file usage a bit. * Compile the parser as C++.
2006-09-04 23:06:23 +02:00
};
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
static string showAttrPath(const vector<Symbol> & attrPath)
* Check for duplicate attributes in fixAttrs, rather than doing a separate traversal after parsing. Likewise, check for duplicate pattern variables right away.
2009-05-14 16:29:45 +02:00
{
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
string s;
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
foreach (vector<Symbol>::const_iterator, i, attrPath) {
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
if (!s.empty()) s += '.';
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
s += *i;
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
}
return s;
}
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
static void dupAttr(const vector<Symbol> & attrPath, const Pos & pos, const Pos & prevPos)
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
{
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
throw ParseError(format("attribute `%1%' at %2% already defined at %3%")
% showAttrPath(attrPath) % pos % prevPos);
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
}
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
static void dupAttr(Symbol attr, const Pos & pos, const Pos & prevPos)
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
{
vector<Symbol> attrPath; attrPath.push_back(attr);
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
throw ParseError(format("attribute `%1%' at %2% already defined at %3%")
% showAttrPath(attrPath) % pos % prevPos);
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
}
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
static void addAttr(ExprAttrs * attrs, const vector<Symbol> & attrPath,
Expr * e, const Pos & pos)
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
{
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
unsigned int n = 0;
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
foreach (vector<Symbol>::const_iterator, i, attrPath) {
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
n++;
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
ExprAttrs::Attrs::iterator j = attrs->attrs.find(*i);
if (j != attrs->attrs.end()) {
ExprAttrs * attrs2 = dynamic_cast<ExprAttrs *>(j->second.first);
if (!attrs2 || n == attrPath.size()) dupAttr(attrPath, pos, j->second.second);
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
attrs = attrs2;
} else {
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
if (attrs->attrNames.find(*i) != attrs->attrNames.end())
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
dupAttr(attrPath, pos, attrs->attrNames[*i]);
attrs->attrNames[*i] = pos;
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
if (n == attrPath.size())
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
attrs->attrs[*i] = ExprAttrs::Attr(e, pos);
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
else {
ExprAttrs * nested = new ExprAttrs;
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
attrs->attrs[*i] = ExprAttrs::Attr(nested, pos);
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
attrs = nested;
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
}
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
}
* Check for duplicate attributes in fixAttrs, rather than doing a separate traversal after parsing. Likewise, check for duplicate pattern variables right away.
2009-05-14 16:29:45 +02:00
}
}
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
static void addFormal(const Pos & pos, Formals * formals, const Formal & formal)
* Check for duplicate attributes in fixAttrs, rather than doing a separate traversal after parsing. Likewise, check for duplicate pattern variables right away.
2009-05-14 16:29:45 +02:00
{
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
if (formals->argNames.find(formal.name) != formals->argNames.end())
throw ParseError(format("duplicate formal function argument `%1%' at %2%")
% formal.name % pos);
formals->formals.push_front(formal);
formals->argNames.insert(formal.name);
* Check for duplicate attributes in fixAttrs, rather than doing a separate traversal after parsing. Likewise, check for duplicate pattern variables right away.
2009-05-14 16:29:45 +02:00
}
* Indented strings.
2010-04-13 00:03:27 +02:00
static Expr * stripIndentation(vector<Expr *> & es)
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
{
* Indented strings.
2010-04-13 00:03:27 +02:00
if (es.empty()) return new ExprString("");
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
/* Figure out the minimum indentation. Note that by design
whitespace-only final lines are not taken into account. (So
the " " in "\n ''" is ignored, but the " " in "\n foo''" is.) */
bool atStartOfLine = true; /* = seen only whitespace in the current line */
unsigned int minIndent = 1000000;
unsigned int curIndent = 0;
* Indented strings.
2010-04-13 00:03:27 +02:00
foreach (vector<Expr *>::iterator, i, es) {
ExprIndStr * e = dynamic_cast<ExprIndStr *>(*i);
if (!e) {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
/* Anti-quotations end the current start-of-line whitespace. */
if (atStartOfLine) {
atStartOfLine = false;
if (curIndent < minIndent) minIndent = curIndent;
}
continue;
}
* Indented strings.
2010-04-13 00:03:27 +02:00
for (unsigned int j = 0; j < e->s.size(); ++j) {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
if (atStartOfLine) {
* Indented strings.
2010-04-13 00:03:27 +02:00
if (e->s[j] == ' ')
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
curIndent++;
* Indented strings.
2010-04-13 00:03:27 +02:00
else if (e->s[j] == '\n') {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
/* Empty line, doesn't influence minimum
indentation. */
curIndent = 0;
} else {
atStartOfLine = false;
if (curIndent < minIndent) minIndent = curIndent;
}
* Indented strings.
2010-04-13 00:03:27 +02:00
} else if (e->s[j] == '\n') {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
atStartOfLine = true;
curIndent = 0;
}
}
}
/* Strip spaces from each line. */
* Indented strings.
2010-04-13 00:03:27 +02:00
vector<Expr *> * es2 = new vector<Expr *>;
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
atStartOfLine = true;
unsigned int curDropped = 0;
* Indented strings.
2010-04-13 00:03:27 +02:00
unsigned int n = es.size();
for (vector<Expr *>::iterator i = es.begin(); i != es.end(); ++i, --n) {
ExprIndStr * e = dynamic_cast<ExprIndStr *>(*i);
if (!e) {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
atStartOfLine = false;
curDropped = 0;
* Indented strings.
2010-04-13 00:03:27 +02:00
es2->push_back(*i);
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
continue;
}
string s2;
* Indented strings.
2010-04-13 00:03:27 +02:00
for (unsigned int j = 0; j < e->s.size(); ++j) {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
if (atStartOfLine) {
* Indented strings.
2010-04-13 00:03:27 +02:00
if (e->s[j] == ' ') {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
if (curDropped++ >= minIndent)
* Indented strings.
2010-04-13 00:03:27 +02:00
s2 += e->s[j];
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
}
* Indented strings.
2010-04-13 00:03:27 +02:00
else if (e->s[j] == '\n') {
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
curDropped = 0;
* Indented strings.
2010-04-13 00:03:27 +02:00
s2 += e->s[j];
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
} else {
atStartOfLine = false;
curDropped = 0;
* Indented strings.
2010-04-13 00:03:27 +02:00
s2 += e->s[j];
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
}
} else {
* Indented strings.
2010-04-13 00:03:27 +02:00
s2 += e->s[j];
if (e->s[j] == '\n') atStartOfLine = true;
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
}
}
/* Remove the last line if it is empty and consists only of
spaces. */
if (n == 1) {
* Fix a few "comparison is always false/true due to limited range of data type" warnings on 64-bit platforms. The one in parser.y is likely to be a real bug.
2009-04-16 14:03:17 +02:00
string::size_type p = s2.find_last_of('\n');
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
if (p != string::npos && s2.find_first_not_of(' ', p + 1) == string::npos)
s2 = string(s2, 0, p + 1);
}
* Indented strings.
2010-04-13 00:03:27 +02:00
es2->push_back(new ExprString(s2));
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
}
* Indented strings.
2010-04-13 00:03:27 +02:00
return new ExprConcatStrings(es2);
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
}
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
void backToString(yyscan_t scanner);
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
void backToIndString(yyscan_t scanner);
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
static Pos makeCurPos(const YYLTYPE & loc, ParseData * data)
* When something goes wrong in the evaluation of a Nix expression, print a nice backtrace of the stack, rather than vomiting a gigantic (and useless) aterm on the screen. Example: error: while evaluating file `.../pkgs/system/test.nix': while evaluating attribute `subversion' at `.../pkgs/system/all-packages-generic.nix', line 533: while evaluating function at `.../pkgs/applications/version-management/subversion/default.nix', line 1: assertion failed at `.../pkgs/applications/version-management/subversion/default.nix', line 13 Since the Nix expression language is lazy, the trace may be misleading. The purpose is to provide a hint as to the location of the problem.
2004-04-06 00:27:41 +02:00
{
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
return Pos(data->path, loc.first_line, loc.first_column);
* When something goes wrong in the evaluation of a Nix expression, print a nice backtrace of the stack, rather than vomiting a gigantic (and useless) aterm on the screen. Example: error: while evaluating file `.../pkgs/system/test.nix': while evaluating attribute `subversion' at `.../pkgs/system/all-packages-generic.nix', line 533: while evaluating function at `.../pkgs/applications/version-management/subversion/default.nix', line 1: assertion failed at `.../pkgs/applications/version-management/subversion/default.nix', line 13 Since the Nix expression language is lazy, the trace may be misleading. The purpose is to provide a hint as to the location of the problem.
2004-04-06 00:27:41 +02:00
}
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
#define CUR_POS makeCurPos(*yylocp, data)
* Override YYMALLOC and YYFREE so that we can call AT[un]protectMemory on the Bison parse stack. Otherwise, a garbage collect during parsing could lead to a crash.
2006-02-13 14:09:23 +01:00
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
}
* Hack for Bison 2.3 compatability.
2006-10-02 16:43:15 +02:00
void yyerror(YYLTYPE * loc, yyscan_t scanner, ParseData * data, const char * error)
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
{
* More missing constructs.
2010-04-12 23:21:24 +02:00
data->error = (format("%1%, at %2%")
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
% error % makeCurPos(*loc, data)).str();
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
}
* Hack for Bison 2.3 compatability.
2006-10-02 16:43:15 +02:00
#endif
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%}
%union {
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
nix::Expr * e;
nix::ExprList * list;
nix::ExprAttrs * attrs;
nix::Formals * formals;
nix::Formal * formal;
* Don't use ATerms to represent integers in the lexer.
2010-04-12 12:38:18 +02:00
int n;
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
char * id; // !!! -> Symbol
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
char * path;
char * uri;
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
std::vector<nix::Symbol> * ids;
* More missing constructs.
2010-04-12 23:21:24 +02:00
std::vector<nix::Expr *> * string_parts;
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
}
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
%type <e> start expr expr_function expr_if expr_op
%type <e> expr_app expr_select expr_simple
%type <list> expr_list
%type <attrs> binds
* Added an experimental feature suggested by Andres: ellipses ("...") in attribute set pattern matches. This allows defining a function that takes *at least* the listed attributes, while ignoring additional attributes. For instance, {stdenv, fetchurl, fuse, ...}: stdenv.mkDerivation { ... }; defines a function that requires an attribute set that contains the specified attributes but ignores others. The main advantage is that we can then write in all-packages.nix aefs = import ../bla/aefs pkgs; instead of aefs = import ../bla/aefs { inherit stdenv fetchurl fuse; }; This saves a lot of typing (not to mention not having to update all-packages.nix with purely mechanical changes). It saves as much typing as the "args: with args;" style, but has the advantage that the function arguments are properly declared (not implicit in what the body of the "with" uses).
2008-08-14 16:00:44 +02:00
%type <formals> formals
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
%type <formal> formal
* Indented strings.
2010-04-13 00:03:27 +02:00
%type <ids> ids attrpath
%type <string_parts> string_parts ind_string_parts
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
%token <id> ID ATTRPATH
* More missing constructs.
2010-04-12 23:21:24 +02:00
%token <e> STR IND_STR
* Don't use ATerms to represent integers in the lexer.
2010-04-12 12:38:18 +02:00
%token <n> INT
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
%token <path> PATH
%token <uri> URI
* Finally, a real "let" syntax: `let x = ...; ... z = ...; in ...'.
2006-10-02 17:52:44 +02:00
%token IF THEN ELSE ASSERT WITH LET IN REC INHERIT EQ NEQ AND OR IMPL
* String interpolation. Expressions like "--with-freetype2-library=" + freetype + "/lib" can now be written as "--with-freetype2-library=${freetype}/lib" An arbitrary expression can be enclosed within ${...}, not just identifiers. * Escaping in string literals: \n, \r, \t interpreted as in C, any other character following \ is interpreted as-is. * Newlines are now allowed in string literals.
2006-05-01 16:01:47 +02:00
%token DOLLAR_CURLY /* == ${ */
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
%token IND_STRING_OPEN IND_STRING_CLOSE
* Added an experimental feature suggested by Andres: ellipses ("...") in attribute set pattern matches. This allows defining a function that takes *at least* the listed attributes, while ignoring additional attributes. For instance, {stdenv, fetchurl, fuse, ...}: stdenv.mkDerivation { ... }; defines a function that requires an attribute set that contains the specified attributes but ignores others. The main advantage is that we can then write in all-packages.nix aefs = import ../bla/aefs pkgs; instead of aefs = import ../bla/aefs { inherit stdenv fetchurl fuse; }; This saves a lot of typing (not to mention not having to update all-packages.nix with purely mechanical changes). It saves as much typing as the "args: with args;" style, but has the advantage that the function arguments are properly declared (not implicit in what the body of the "with" uses).
2008-08-14 16:00:44 +02:00
%token ELLIPSIS
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%nonassoc IMPL
%left OR
%left AND
%nonassoc EQ NEQ
* An attribute set update operator (//). E.g., {x=1; y=2; z=3;} // {y=4;} => {x=1; y=4; z=3;}
2004-02-04 17:49:51 +01:00
%right UPDATE
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%left NEG
* String/path concatenation operator (`+').
2004-10-26 19:01:35 +02:00
%left '+'
* List concatenation must be right-associative for efficiency.
2005-09-14 13:41:59 +02:00
%right CONCAT
* Added an operator `?' to test for attribute existence, e.g., `attrs ? x' yields true iff `attrs' has an attribute named `x'.
2004-03-28 23:15:01 +02:00
%nonassoc '?'
* Added an operator `~' to select paths within a derivation. E.g., {stdenv, bash}: derivation { builder = bash ~ /bin/sh; args = ["-e" "-x" ./builder.sh]; ... } Here the attribute `builder' will evaluate to, e.g., `/nix/store/1234abcd...-bash-2.0.1/bin/sh'.
2004-03-28 22:58:28 +02:00
%nonassoc '~'
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
%%
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
start: expr { data->result = $1; };
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
expr: expr_function;
expr_function
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
: ID ':' expr_function
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
{ $$ = new ExprLambda(CUR_POS, data->symbols.create($1), false, 0, $3); }
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| '{' formals '}' ':' expr_function
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
{ $$ = new ExprLambda(CUR_POS, data->symbols.create(""), true, $2, $5); }
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| '{' formals '}' '@' ID ':' expr_function
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
{ $$ = new ExprLambda(CUR_POS, data->symbols.create($5), true, $2, $7); }
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| ID '@' '{' formals '}' ':' expr_function
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
{ $$ = new ExprLambda(CUR_POS, data->symbols.create($1), true, $4, $7); }
* More missing constructs.
2010-04-12 23:21:24 +02:00
| ASSERT expr ';' expr_function
{ $$ = new ExprAssert(CUR_POS, $2, $4); }
* New language feature: with expressions. The expression `with E1; E2' evaluates to E2 with all bindings in the attribute set E1 substituted. E.g., with {x = 123;}; x evaluates to 123. That is, the attribute set E1 is in scope in E2. This is particularly useful when importing files containing lots definitions. E.g., instead of let { inherit (import ./foo.nix) a b c d e f; body = ... a ... f ...; } we can now say with import ./foo.nix; ... a ... f ... I.e., we don't have to say what variables should be brought into scope.
2004-10-25 18:54:56 +02:00
| WITH expr ';' expr_function
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
{ $$ = new ExprWith(CUR_POS, $2, $4); }
* Finally, a real "let" syntax: `let x = ...; ... z = ...; in ...'.
2006-10-02 17:52:44 +02:00
| LET binds IN expr_function
* Evaluate lets directly (i.e. without desugaring to `rec { attrs...; <let-body> = e; }.<let-body>). This prevents the unnecessary allocation of an attribute set.
2010-04-13 15:42:25 +02:00
{ $$ = new ExprLet($2, $4); }
* Resolve an ambiguity between ifs and attribute selection, e.g., `if b then x else y.z'.
2004-02-19 14:11:12 +01:00
| expr_if
;
expr_if
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
: IF expr THEN expr ELSE expr { $$ = new ExprIf($2, $4, $6); }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
| expr_op
;
expr_op
* More missing constructs.
2010-04-12 23:21:24 +02:00
: '!' expr_op %prec NEG { $$ = new ExprOpNot($2); }
| expr_op EQ expr_op { $$ = new ExprOpEq($1, $3); }
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| expr_op NEQ expr_op { $$ = new ExprOpNEq($1, $3); }
| expr_op AND expr_op { $$ = new ExprOpAnd($1, $3); }
| expr_op OR expr_op { $$ = new ExprOpOr($1, $3); }
| expr_op IMPL expr_op { $$ = new ExprOpImpl($1, $3); }
| expr_op UPDATE expr_op { $$ = new ExprOpUpdate($1, $3); }
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
| expr_op '?' ID { $$ = new ExprOpHasAttr($1, data->symbols.create($3)); }
* More missing constructs.
2010-04-12 23:21:24 +02:00
| expr_op '+' expr_op
{ vector<Expr *> * l = new vector<Expr *>;
l->push_back($1);
l->push_back($3);
$$ = new ExprConcatStrings(l);
}
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| expr_op CONCAT expr_op { $$ = new ExprOpConcatLists($1, $3); }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
| expr_app
;
expr_app
: expr_app expr_select
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
{ $$ = new ExprApp($1, $2); }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
| expr_select { $$ = $1; }
;
expr_select
: expr_select '.' ID
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
{ $$ = new ExprSelect($1, data->symbols.create($3)); }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
| expr_simple { $$ = $1; }
;
expr_simple
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
: ID { $$ = new ExprVar(data->symbols.create($1)); }
* More missing constructs.
2010-04-12 23:21:24 +02:00
| INT { $$ = new ExprInt($1); }
* String interpolation. Expressions like "--with-freetype2-library=" + freetype + "/lib" can now be written as "--with-freetype2-library=${freetype}/lib" An arbitrary expression can be enclosed within ${...}, not just identifiers. * Escaping in string literals: \n, \r, \t interpreted as in C, any other character following \ is interpreted as-is. * Newlines are now allowed in string literals.
2006-05-01 16:01:47 +02:00
| '"' string_parts '"' {
* More missing constructs.
2010-04-12 23:21:24 +02:00
/* For efficiency, and to simplify parse trees a bit. */
if ($2->empty()) $$ = new ExprString("");
else if ($2->size() == 1) $$ = $2->front();
else $$ = new ExprConcatStrings($2);
* String interpolation. Expressions like "--with-freetype2-library=" + freetype + "/lib" can now be written as "--with-freetype2-library=${freetype}/lib" An arbitrary expression can be enclosed within ${...}, not just identifiers. * Escaping in string literals: \n, \r, \t interpreted as in C, any other character following \ is interpreted as-is. * Newlines are now allowed in string literals.
2006-05-01 16:01:47 +02:00
}
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
| IND_STRING_OPEN ind_string_parts IND_STRING_CLOSE {
* Indented strings.
2010-04-13 00:03:27 +02:00
$$ = stripIndentation(*$2);
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
}
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| PATH { $$ = new ExprPath(absPath($1, data->basePath)); }
| URI { $$ = new ExprString($1); }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
| '(' expr ')' { $$ = $2; }
* Added syntactic sugar to the construction of attribute sets to `inherit' variables from the surrounding lexical scope. E.g., {stdenv, libfoo}: derivation { builder = ./bla; inherit stdenv libfoo; xyzzy = 1; } is equivalent to {stdenv, libfoo}: derivation { builder = ./bla; stdenv = stdenv; libfoo = libfoo; xyzzy = 1; } Note that for mutually recursive attribute set definitions (`rec {...}'), this also works, that is, `rec {inherit x;}' is equivalent to `let {fresh = x; body = rec {x = fresh;};}', *not* `rec {x = x}'.
2004-02-02 22:39:33 +01:00
/* Let expressions `let {..., body = ...}' are just desugared
* More missing constructs.
2010-04-12 23:21:24 +02:00
into `(rec {..., body = ...}).body'. */
* Added syntactic sugar to the construction of attribute sets to `inherit' variables from the surrounding lexical scope. E.g., {stdenv, libfoo}: derivation { builder = ./bla; inherit stdenv libfoo; xyzzy = 1; } is equivalent to {stdenv, libfoo}: derivation { builder = ./bla; stdenv = stdenv; libfoo = libfoo; xyzzy = 1; } Note that for mutually recursive attribute set definitions (`rec {...}'), this also works, that is, `rec {inherit x;}' is equivalent to `let {fresh = x; body = rec {x = fresh;};}', *not* `rec {x = x}'.
2004-02-02 22:39:33 +01:00
| LET '{' binds '}'
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
{ $3->recursive = true; $$ = new ExprSelect($3, data->symbols.create("body")); }
* Added syntactic sugar to the construction of attribute sets to `inherit' variables from the surrounding lexical scope. E.g., {stdenv, libfoo}: derivation { builder = ./bla; inherit stdenv libfoo; xyzzy = 1; } is equivalent to {stdenv, libfoo}: derivation { builder = ./bla; stdenv = stdenv; libfoo = libfoo; xyzzy = 1; } Note that for mutually recursive attribute set definitions (`rec {...}'), this also works, that is, `rec {inherit x;}' is equivalent to `let {fresh = x; body = rec {x = fresh;};}', *not* `rec {x = x}'.
2004-02-02 22:39:33 +01:00
| REC '{' binds '}'
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
{ $3->recursive = true; $$ = $3; }
* Added syntactic sugar to the construction of attribute sets to `inherit' variables from the surrounding lexical scope. E.g., {stdenv, libfoo}: derivation { builder = ./bla; inherit stdenv libfoo; xyzzy = 1; } is equivalent to {stdenv, libfoo}: derivation { builder = ./bla; stdenv = stdenv; libfoo = libfoo; xyzzy = 1; } Note that for mutually recursive attribute set definitions (`rec {...}'), this also works, that is, `rec {inherit x;}' is equivalent to `let {fresh = x; body = rec {x = fresh;};}', *not* `rec {x = x}'.
2004-02-02 22:39:33 +01:00
| '{' binds '}'
* Finished the ATerm-less parser.
2010-04-13 01:33:23 +02:00
{ $$ = $2; }
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| '[' expr_list ']' { $$ = $2; }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
;
* String interpolation. Expressions like "--with-freetype2-library=" + freetype + "/lib" can now be written as "--with-freetype2-library=${freetype}/lib" An arbitrary expression can be enclosed within ${...}, not just identifiers. * Escaping in string literals: \n, \r, \t interpreted as in C, any other character following \ is interpreted as-is. * Newlines are now allowed in string literals.
2006-05-01 16:01:47 +02:00
string_parts
* More missing constructs.
2010-04-12 23:21:24 +02:00
: string_parts STR { $$ = $1; $1->push_back($2); }
| string_parts DOLLAR_CURLY expr '}' { backToString(scanner); $$ = $1; $1->push_back($3); }
| { $$ = new vector<Expr *>; }
* String interpolation. Expressions like "--with-freetype2-library=" + freetype + "/lib" can now be written as "--with-freetype2-library=${freetype}/lib" An arbitrary expression can be enclosed within ${...}, not just identifiers. * Escaping in string literals: \n, \r, \t interpreted as in C, any other character following \ is interpreted as-is. * Newlines are now allowed in string literals.
2006-05-01 16:01:47 +02:00
;
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
ind_string_parts
* Indented strings.
2010-04-13 00:03:27 +02:00
: ind_string_parts IND_STR { $$ = $1; $1->push_back($2); }
| ind_string_parts DOLLAR_CURLY expr '}' { backToIndString(scanner); $$ = $1; $1->push_back($3); }
| { $$ = new vector<Expr *>; }
* Added a new kind of multi-line string literal delimited by two single quotes. Example (from NixOS): job = '' start on network-interfaces start script rm -f /var/run/opengl-driver ${if videoDriver == "nvidia" then "ln -sf ${nvidiaDrivers} /var/run/opengl-driver" else if cfg.driSupport then "ln -sf ${mesa} /var/run/opengl-driver" else "" } rm -f /var/log/slim.log end script ''; This style has two big advantages: - \, ' and " aren't special, only '' and ${. So you get a lot less escaping in shell scripts / configuration files in Nixpkgs/NixOS. The delimiter '' is rare in scripts (and can usually be written as ""). ${ is also fairly rare. Other delimiters such as <<...>>, {{...}} and <|...|> were also considered but this one appears to have the fewest drawbacks (thanks Martin). - Indentation is intelligently stripped so that multi-line strings can follow the nesting structure of the containing Nix expression. E.g. in the example above 6 spaces are stripped from the start of each line. This prevents unnecessary indentation in generated files (which sometimes even breaks things). See tests/lang/eval-okay-ind-string.nix for some examples.
2007-11-30 17:48:45 +01:00
;
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
binds
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
: binds attrpath '=' expr ';' { $$ = $1; addAttr($$, *$2, $4, makeCurPos(@2, data)); }
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| binds INHERIT ids ';'
{ $$ = $1;
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
foreach (vector<Symbol>::iterator, i, *$3) {
if ($$->attrNames.find(*i) != $$->attrNames.end())
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
dupAttr(*i, makeCurPos(@3, data), $$->attrNames[*i]);
* Store position info for inherited attributes.
2010-05-07 14:43:57 +02:00
Pos pos = makeCurPos(@3, data);
$$->inherited.push_back(ExprAttrs::Inherited(*i, pos));
$$->attrNames[*i] = pos;
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
}
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
}
| binds INHERIT '(' expr ')' ids ';'
{ $$ = $1;
/* !!! Should ensure sharing of the expression in $4. */
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
foreach (vector<Symbol>::iterator, i, *$6) {
if ($$->attrNames.find(*i) != $$->attrNames.end())
* Store attribute positions in the AST and report duplicate attribute errors with position info. * For all positions, use the position of the first character of the first token, rather than the last character of the first token plus one.
2010-05-06 18:46:48 +02:00
dupAttr(*i, makeCurPos(@6, data), $$->attrNames[*i]);
$$->attrs[*i] = ExprAttrs::Attr(new ExprSelect($4, *i), makeCurPos(@6, data));
$$->attrNames[*i] = makeCurPos(@6, data);
}}
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
| { $$ = new ExprAttrs; }
* Extended the `inherit' syntax to optionally select attributes from other attribute sets, rather than the current scope. E.g., {inherit (pkgs) gcc binutils;} is equivalent to {gcc = pkgs.gcc; binutils = pkgs.binutils;} I am not so happy about the syntax.
2004-02-04 18:23:26 +01:00
;
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
ids
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
: ids ID { $$ = $1; $1->push_back(data->symbols.create($2)); /* !!! dangerous */ }
| { $$ = new vector<Symbol>; }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
;
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
attrpath
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
: attrpath '.' ID { $$ = $1; $1->push_back(data->symbols.create($3)); }
| ID { $$ = new vector<Symbol>; $$->push_back(data->symbols.create($1)); }
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
;
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
expr_list
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
: expr_list expr_select { $$ = $1; $1->elems.push_back($2); /* !!! dangerous */ }
| { $$ = new ExprList; }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
;
formals
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
: formal ',' formals
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
{ $$ = $3; addFormal(CUR_POS, $$, *$1); }
* Added an experimental feature suggested by Andres: ellipses ("...") in attribute set pattern matches. This allows defining a function that takes *at least* the listed attributes, while ignoring additional attributes. For instance, {stdenv, fetchurl, fuse, ...}: stdenv.mkDerivation { ... }; defines a function that requires an attribute set that contains the specified attributes but ignores others. The main advantage is that we can then write in all-packages.nix aefs = import ../bla/aefs pkgs; instead of aefs = import ../bla/aefs { inherit stdenv fetchurl fuse; }; This saves a lot of typing (not to mention not having to update all-packages.nix with purely mechanical changes). It saves as much typing as the "args: with args;" style, but has the advantage that the function arguments are properly declared (not implicit in what the body of the "with" uses).
2008-08-14 16:00:44 +02:00
| formal
* Check for duplicate attribute names / function arguments. `make check' now succeeds :-) * An attribute set such as `{ foo = { enable = true; }; foo.port = 23; }' now parses. It was previously rejected, but I'm too lazy to implement the check. (The only reason to reject it is that the reverse, `{ foo.port = 23; foo = { enable = true; }; }', is rejected, which is kind of ugly.)
2010-04-22 13:02:24 +02:00
{ $$ = new Formals; addFormal(CUR_POS, $$, *$1); $$->ellipsis = false; }
* Added an experimental feature suggested by Andres: ellipses ("...") in attribute set pattern matches. This allows defining a function that takes *at least* the listed attributes, while ignoring additional attributes. For instance, {stdenv, fetchurl, fuse, ...}: stdenv.mkDerivation { ... }; defines a function that requires an attribute set that contains the specified attributes but ignores others. The main advantage is that we can then write in all-packages.nix aefs = import ../bla/aefs pkgs; instead of aefs = import ../bla/aefs { inherit stdenv fetchurl fuse; }; This saves a lot of typing (not to mention not having to update all-packages.nix with purely mechanical changes). It saves as much typing as the "args: with args;" style, but has the advantage that the function arguments are properly declared (not implicit in what the body of the "with" uses).
2008-08-14 16:00:44 +02:00
|
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
{ $$ = new Formals; $$->ellipsis = false; }
* Added an experimental feature suggested by Andres: ellipses ("...") in attribute set pattern matches. This allows defining a function that takes *at least* the listed attributes, while ignoring additional attributes. For instance, {stdenv, fetchurl, fuse, ...}: stdenv.mkDerivation { ... }; defines a function that requires an attribute set that contains the specified attributes but ignores others. The main advantage is that we can then write in all-packages.nix aefs = import ../bla/aefs pkgs; instead of aefs = import ../bla/aefs { inherit stdenv fetchurl fuse; }; This saves a lot of typing (not to mention not having to update all-packages.nix with purely mechanical changes). It saves as much typing as the "args: with args;" style, but has the advantage that the function arguments are properly declared (not implicit in what the body of the "with" uses).
2008-08-14 16:00:44 +02:00
| ELLIPSIS
* Don't use ATerms for the abstract syntax trees anymore. Not finished yet.
2010-04-12 20:30:11 +02:00
{ $$ = new Formals; $$->ellipsis = true; }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
;
formal
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
: ID { $$ = new Formal(data->symbols.create($1), 0); }
| ID '?' expr { $$ = new Formal(data->symbols.create($1), $3); }
* Replaced the SDF parser by a substantially faster Bison/Flex parser (roughly 80x faster). The absolutely latest version of Bison (1.875c) is required for reentrant GLR support, as well as a recent version of Flex (say, 2.5.31). Note that most Unix distributions ship with the prehistoric Flex 2.5.4, which doesn't support reentrancy.
2004-01-30 16:21:42 +01:00
;
%%
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
#include <eval.hh>
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
namespace nix {
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
static Expr * parse(EvalState & state, const char * text,
const Path & path, const Path & basePath)
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
{
yyscan_t scanner;
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
ParseData data(state.symbols);
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
data.basePath = basePath;
data.path = path;
yylex_init(&scanner);
yy_scan_string(text, scanner);
int res = yyparse(scanner, &data);
yylex_destroy(scanner);
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
if (res) throw ParseError(data.error);
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
try {
* Fix builtins.
2010-04-15 00:59:39 +02:00
data.result->bindVars(state.staticBaseEnv);
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
} catch (Error & e) {
* Some syntactic sugar for attribute sets: allow {x.y.z = ...;} as a shorthand for {x = {y = {z = ...;};};}. This is especially useful for NixOS configuration files, e.g. { services = { sshd = { enable = true; port = 2022; }; }; } can now be written as { services.sshd.enable = true; services.sshd.port = 2022; } However, it is currently not permitted to write { services.sshd = {enable = true;}; services.sshd.port = 2022; } as this is considered a duplicate definition of `services.sshd'.
2009-05-15 14:35:23 +02:00
throw ParseError(format("%1%, in `%2%'") % e.msg() % path);
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
}
return data.result;
}
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
Expr * parseExprFromFile(EvalState & state, Path path)
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
{
assert(path[0] == '/');
/* If `path' is a symlink, follow it. This is so that relative
path references work. */
struct stat st;
* Handle multiple indirect symlinks when loading a Nix expression.
2007-01-15 15:50:25 +01:00
while (true) {
if (lstat(path.c_str(), &st))
throw SysError(format("getting status of `%1%'") % path);
if (!S_ISLNK(st.st_mode)) break;
path = absPath(readLink(path), dirOf(path));
}
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
/* If `path' refers to a directory, append `/default.nix'. */
if (S_ISDIR(st.st_mode))
path = canonPath(path + "/default.nix");
* Don't allocate input files on the stack.
2007-08-07 17:00:13 +02:00
/* Read and parse the input file. */
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
return parse(state, readFile(path).c_str(), path, dirOf(path));
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
}
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
Expr * parseExprFromString(EvalState & state,
const string & s, const Path & basePath)
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
{
* Use a symbol table to represent identifiers and attribute names efficiently. The symbol table ensures that there is only one copy of each symbol, thus allowing symbols to be compared efficiently using a pointer equality test.
2010-04-13 14:25:42 +02:00
return parse(state, s.c_str(), "(string)", basePath);
* Compile the lexer as C++ code. Remove all the redundant C/C++ marshalling code.
2006-09-04 23:36:15 +02:00
}
}
Reference in a new issue Copy permalink