luci-base: add support for plural translations and contexts in Lua api

- Introduce a new luci.template.parser.ntranslate() function which
   takes a count, a singular and a plural translation string as well
   as an optional context argument and returns the appropriate,
   language specific plural translation.

 - Introduce an optional translation context argument in the existing
   luci.template.parser.translate() function

 - Support translation contexts in LuCI template directives.
   Translation messages are split on the first unescaped pipe
   character and the reamining string after the pipe is treated
   as context.

Examples:

 - `string.format(p.ntranslate(n, "1 apple", "%d apples"), n)` will
   return an appropriate plural translation for the given amount.

 - `translate("Load", "The system load")` will return an appropiate
   translation for `Load`, using `The system load` as disambiguation
   context (a `msgctxt` directive in *.po files).

 - Likewise `<%:Load|The system load%>` will translate the word
   `Load` while using the remainder of the string as context.

 - To use pipes in translations strings literally, they must be
   escaped: `<%:Use the "\|" character%>` will translate the literal
   string `Use the "|" character`.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This commit is contained in:
Jo-Philipp Wich 2020-01-23 21:53:26 +01:00
parent 6f6f3e84ca
commit 9939fc5a26
10 changed files with 6319 additions and 17 deletions

3
.gitignore vendored
View file

@ -6,3 +6,6 @@ dist/
*.po~
modules/luci-base/src/po2lmo
modules/luci-base/src/jsmin
modules/luci-base/src/contrib/lemon
modules/luci-base/src/plural_formula.c
modules/luci-base/src/plural_formula.h

View file

@ -124,14 +124,24 @@ sub preprocess_htm($$) {
'=' => '(%s)',
'_' => 'translate([==[%s]==])',
':' => 'translate([==[%s]==])',
'+' => 'include([==[%s]==)',
'+' => 'include([==[%s]==])',
'#' => '--[==[%s]==]',
'' => '%s'
};
# Translate the .htm source into a valid Lua source using bracket quotes
# to avoid the need for complex escaping.
$source =~ s|<%-?([=_:+#]?)(.*?)-?%>|sprintf "]==]; $sub->{$1}; [==[", $2|sge;
$source =~ s!<%-?([=_:+#]?)(.*?)-?%>!
my $t = $1;
my $s = $2;
# Split translation expressions on first non-escaped pipe.
if ($t eq ':' || $t eq '_') {
$s =~ s/^((?:[^\|\\]|\\.)*)\|(.*)$/$1]==],[==[$2/;
}
sprintf "]==]; $sub->{$t}; [==[", $s
!sge;
# Discover expressions like "lng.translate(...)" or "luci.i18n.translate(...)"
# and return them as extra keyword so that xgettext recognizes such expressions

View file

@ -1,16 +1,24 @@
%.o: %.c
$(CC) $(CPPFLAGS) $(CFLAGS) $(FPIC) -c -o $@ $<
$(CC) $(CPPFLAGS) $(CFLAGS) $(FPIC) -DNDEBUG -c -o $@ $<
contrib/lemon: contrib/lemon.c contrib/lempar.c
cc -o contrib/lemon $<
plural_formula.c: plural_formula.y contrib/lemon
./contrib/lemon -q $<
template_lmo.c: plural_formula.c
clean:
rm -f po2lmo parser.so version.lua *.o
rm -f contrib/lemon po2lmo parser.so version.lua plural_formula.c plural_formula.h *.o
jsmin: jsmin.o
$(CC) $(LDFLAGS) -o $@ $^
po2lmo: po2lmo.o template_lmo.o
po2lmo: po2lmo.o template_lmo.o plural_formula.o
$(CC) $(LDFLAGS) -o $@ $^
parser.so: template_parser.o template_utils.o template_lmo.o template_lualib.o
parser.so: template_parser.o template_utils.o template_lmo.o template_lualib.o plural_formula.o
$(CC) $(LDFLAGS) -shared -o $@ $^
version.lua:

File diff suppressed because it is too large Load diff

View file

@ -0,0 +1,851 @@
/* Driver template for the LEMON parser generator.
** The author disclaims copyright to this source code.
*/
/* First off, code is included that follows the "include" declaration
** in the input grammar file. */
#include <stdio.h>
%%
/* Next is all token values, in a form suitable for use by makeheaders.
** This section will be null unless lemon is run with the -m switch.
*/
/*
** These constants (all generated automatically by the parser generator)
** specify the various kinds of tokens (terminals) that the parser
** understands.
**
** Each symbol here is a terminal symbol in the grammar.
*/
%%
/* Make sure the INTERFACE macro is defined.
*/
#ifndef INTERFACE
# define INTERFACE 1
#endif
/* The next thing included is series of defines which control
** various aspects of the generated parser.
** YYCODETYPE is the data type used for storing terminal
** and nonterminal numbers. "unsigned char" is
** used if there are fewer than 250 terminals
** and nonterminals. "int" is used otherwise.
** YYNOCODE is a number of type YYCODETYPE which corresponds
** to no legal terminal or nonterminal number. This
** number is used to fill in empty slots of the hash
** table.
** YYFALLBACK If defined, this indicates that one or more tokens
** have fall-back values which should be used if the
** original value of the token will not parse.
** YYACTIONTYPE is the data type used for storing terminal
** and nonterminal numbers. "unsigned char" is
** used if there are fewer than 250 rules and
** states combined. "int" is used otherwise.
** ParseTOKENTYPE is the data type used for minor tokens given
** directly to the parser from the tokenizer.
** YYMINORTYPE is the data type used for all minor tokens.
** This is typically a union of many types, one of
** which is ParseTOKENTYPE. The entry in the union
** for base tokens is called "yy0".
** YYSTACKDEPTH is the maximum depth of the parser's stack. If
** zero the stack is dynamically sized using realloc()
** ParseARG_SDECL A static variable declaration for the %extra_argument
** ParseARG_PDECL A parameter declaration for the %extra_argument
** ParseARG_STORE Code to store %extra_argument into yypParser
** ParseARG_FETCH Code to extract %extra_argument from yypParser
** YYNSTATE the combined number of states.
** YYNRULE the number of rules in the grammar
** YYERRORSYMBOL is the code number of the error symbol. If not
** defined, then do no error processing.
*/
%%
#define YY_NO_ACTION (YYNSTATE+YYNRULE+2)
#define YY_ACCEPT_ACTION (YYNSTATE+YYNRULE+1)
#define YY_ERROR_ACTION (YYNSTATE+YYNRULE)
/* The yyzerominor constant is used to initialize instances of
** YYMINORTYPE objects to zero. */
static const YYMINORTYPE yyzerominor = { 0 };
/* Define the yytestcase() macro to be a no-op if is not already defined
** otherwise.
**
** Applications can choose to define yytestcase() in the %include section
** to a macro that can assist in verifying code coverage. For production
** code the yytestcase() macro should be turned off. But it is useful
** for testing.
*/
#ifndef yytestcase
# define yytestcase(X)
#endif
/* Next are the tables used to determine what action to take based on the
** current state and lookahead token. These tables are used to implement
** functions that take a state number and lookahead value and return an
** action integer.
**
** Suppose the action integer is N. Then the action is determined as
** follows
**
** 0 <= N < YYNSTATE Shift N. That is, push the lookahead
** token onto the stack and goto state N.
**
** YYNSTATE <= N < YYNSTATE+YYNRULE Reduce by rule N-YYNSTATE.
**
** N == YYNSTATE+YYNRULE A syntax error has occurred.
**
** N == YYNSTATE+YYNRULE+1 The parser accepts its input.
**
** N == YYNSTATE+YYNRULE+2 No such action. Denotes unused
** slots in the yy_action[] table.
**
** The action table is constructed as a single large table named yy_action[].
** Given state S and lookahead X, the action is computed as
**
** yy_action[ yy_shift_ofst[S] + X ]
**
** If the index value yy_shift_ofst[S]+X is out of range or if the value
** yy_lookahead[yy_shift_ofst[S]+X] is not equal to X or if yy_shift_ofst[S]
** is equal to YY_SHIFT_USE_DFLT, it means that the action is not in the table
** and that yy_default[S] should be used instead.
**
** The formula above is for computing the action when the lookahead is
** a terminal symbol. If the lookahead is a non-terminal (as occurs after
** a reduce action) then the yy_reduce_ofst[] array is used in place of
** the yy_shift_ofst[] array and YY_REDUCE_USE_DFLT is used in place of
** YY_SHIFT_USE_DFLT.
**
** The following are the tables generated in this section:
**
** yy_action[] A single table containing all actions.
** yy_lookahead[] A table containing the lookahead for each entry in
** yy_action. Used to detect hash collisions.
** yy_shift_ofst[] For each state, the offset into yy_action for
** shifting terminals.
** yy_reduce_ofst[] For each state, the offset into yy_action for
** shifting non-terminals after a reduce.
** yy_default[] Default action for each state.
*/
%%
/* The next table maps tokens into fallback tokens. If a construct
** like the following:
**
** %fallback ID X Y Z.
**
** appears in the grammar, then ID becomes a fallback token for X, Y,
** and Z. Whenever one of the tokens X, Y, or Z is input to the parser
** but it does not parse, the type of the token is changed to ID and
** the parse is retried before an error is thrown.
*/
#ifdef YYFALLBACK
static const YYCODETYPE yyFallback[] = {
%%
};
#endif /* YYFALLBACK */
/* The following structure represents a single element of the
** parser's stack. Information stored includes:
**
** + The state number for the parser at this level of the stack.
**
** + The value of the token stored at this level of the stack.
** (In other words, the "major" token.)
**
** + The semantic value stored at this level of the stack. This is
** the information used by the action routines in the grammar.
** It is sometimes called the "minor" token.
*/
struct yyStackEntry {
YYACTIONTYPE stateno; /* The state-number */
YYCODETYPE major; /* The major token value. This is the code
** number for the token at this stack level */
YYMINORTYPE minor; /* The user-supplied minor token value. This
** is the value of the token */
};
typedef struct yyStackEntry yyStackEntry;
/* The state of the parser is completely contained in an instance of
** the following structure */
struct yyParser {
int yyidx; /* Index of top element in stack */
#ifdef YYTRACKMAXSTACKDEPTH
int yyidxMax; /* Maximum value of yyidx */
#endif
int yyerrcnt; /* Shifts left before out of the error */
ParseARG_SDECL /* A place to hold %extra_argument */
#if YYSTACKDEPTH<=0
int yystksz; /* Current side of the stack */
yyStackEntry *yystack; /* The parser's stack */
#else
yyStackEntry yystack[YYSTACKDEPTH]; /* The parser's stack */
#endif
};
typedef struct yyParser yyParser;
#ifndef NDEBUG
#include <stdio.h>
static FILE *yyTraceFILE = 0;
static char *yyTracePrompt = 0;
#endif /* NDEBUG */
#ifndef NDEBUG
/*
** Turn parser tracing on by giving a stream to which to write the trace
** and a prompt to preface each trace message. Tracing is turned off
** by making either argument NULL
**
** Inputs:
** <ul>
** <li> A FILE* to which trace output should be written.
** If NULL, then tracing is turned off.
** <li> A prefix string written at the beginning of every
** line of trace output. If NULL, then tracing is
** turned off.
** </ul>
**
** Outputs:
** None.
*/
void ParseTrace(FILE *TraceFILE, char *zTracePrompt);
void ParseTrace(FILE *TraceFILE, char *zTracePrompt){
yyTraceFILE = TraceFILE;
yyTracePrompt = zTracePrompt;
if( yyTraceFILE==0 ) yyTracePrompt = 0;
else if( yyTracePrompt==0 ) yyTraceFILE = 0;
}
#endif /* NDEBUG */
#ifndef NDEBUG
/* For tracing shifts, the names of all terminals and nonterminals
** are required. The following table supplies these names */
static const char *const yyTokenName[] = {
%%
};
#endif /* NDEBUG */
#ifndef NDEBUG
/* For tracing reduce actions, the names of all rules are required.
*/
static const char *const yyRuleName[] = {
%%
};
#endif /* NDEBUG */
#if YYSTACKDEPTH<=0
/*
** Try to increase the size of the parser stack.
*/
static void yyGrowStack(yyParser *p){
int newSize;
yyStackEntry *pNew;
newSize = p->yystksz*2 + 100;
pNew = realloc(p->yystack, newSize*sizeof(pNew[0]));
if( pNew ){
p->yystack = pNew;
p->yystksz = newSize;
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE,"%sStack grows to %d entries!\n",
yyTracePrompt, p->yystksz);
}
#endif
}
}
#endif
/*
** This function allocates a new parser.
** The only argument is a pointer to a function which works like
** malloc.
**
** Inputs:
** A pointer to the function used to allocate memory.
**
** Outputs:
** A pointer to a parser. This pointer is used in subsequent calls
** to Parse and ParseFree.
*/
void *ParseAlloc(void *(*mallocProc)(size_t)){
yyParser *pParser;
pParser = (yyParser*)(*mallocProc)( (size_t)sizeof(yyParser) );
if( pParser ){
pParser->yyidx = -1;
#ifdef YYTRACKMAXSTACKDEPTH
pParser->yyidxMax = 0;
#endif
#if YYSTACKDEPTH<=0
pParser->yystack = NULL;
pParser->yystksz = 0;
yyGrowStack(pParser);
#endif
}
return pParser;
}
/* The following function deletes the value associated with a
** symbol. The symbol can be either a terminal or nonterminal.
** "yymajor" is the symbol code, and "yypminor" is a pointer to
** the value.
*/
static void yy_destructor(
yyParser *yypParser, /* The parser */
YYCODETYPE yymajor, /* Type code for object to destroy */
YYMINORTYPE *yypminor /* The object to be destroyed */
){
ParseARG_FETCH;
switch( yymajor ){
/* Here is inserted the actions which take place when a
** terminal or non-terminal is destroyed. This can happen
** when the symbol is popped from the stack during a
** reduce or during error processing or when a parser is
** being destroyed before it is finished parsing.
**
** Note: during a reduce, the only symbols destroyed are those
** which appear on the RHS of the rule, but which are not used
** inside the C code.
*/
%%
default: break; /* If no destructor action specified: do nothing */
}
}
/*
** Pop the parser's stack once.
**
** If there is a destructor routine associated with the token which
** is popped from the stack, then call it.
**
** Return the major token number for the symbol popped.
*/
static int yy_pop_parser_stack(yyParser *pParser){
YYCODETYPE yymajor;
yyStackEntry *yytos = &pParser->yystack[pParser->yyidx];
if( pParser->yyidx<0 ) return 0;
#ifndef NDEBUG
if( yyTraceFILE && pParser->yyidx>=0 ){
fprintf(yyTraceFILE,"%sPopping %s\n",
yyTracePrompt,
yyTokenName[yytos->major]);
}
#endif
yymajor = yytos->major;
yy_destructor(pParser, yymajor, &yytos->minor);
pParser->yyidx--;
return yymajor;
}
/*
** Deallocate and destroy a parser. Destructors are all called for
** all stack elements before shutting the parser down.
**
** Inputs:
** <ul>
** <li> A pointer to the parser. This should be a pointer
** obtained from ParseAlloc.
** <li> A pointer to a function used to reclaim memory obtained
** from malloc.
** </ul>
*/
void ParseFree(
void *p, /* The parser to be deleted */
void (*freeProc)(void*) /* Function used to reclaim memory */
){
yyParser *pParser = (yyParser*)p;
if( pParser==0 ) return;
while( pParser->yyidx>=0 ) yy_pop_parser_stack(pParser);
#if YYSTACKDEPTH<=0
free(pParser->yystack);
#endif
(*freeProc)((void*)pParser);
}
/*
** Return the peak depth of the stack for a parser.
*/
#ifdef YYTRACKMAXSTACKDEPTH
int ParseStackPeak(void *p){
yyParser *pParser = (yyParser*)p;
return pParser->yyidxMax;
}
#endif
/*
** Find the appropriate action for a parser given the terminal
** look-ahead token iLookAhead.
**
** If the look-ahead token is YYNOCODE, then check to see if the action is
** independent of the look-ahead. If it is, return the action, otherwise
** return YY_NO_ACTION.
*/
static int yy_find_shift_action(
yyParser *pParser, /* The parser */
YYCODETYPE iLookAhead /* The look-ahead token */
){
int i;
int stateno = pParser->yystack[pParser->yyidx].stateno;
if( stateno>YY_SHIFT_COUNT
|| (i = yy_shift_ofst[stateno])==YY_SHIFT_USE_DFLT ){
return yy_default[stateno];
}
assert( iLookAhead!=YYNOCODE );
i += iLookAhead;
if( i<0 || i>=YY_ACTTAB_COUNT || yy_lookahead[i]!=iLookAhead ){
if( iLookAhead>0 ){
#ifdef YYFALLBACK
YYCODETYPE iFallback; /* Fallback token */
if( iLookAhead<sizeof(yyFallback)/sizeof(yyFallback[0])
&& (iFallback = yyFallback[iLookAhead])!=0 ){
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE, "%sFALLBACK %s => %s\n",
yyTracePrompt, yyTokenName[iLookAhead], yyTokenName[iFallback]);
}
#endif
return yy_find_shift_action(pParser, iFallback);
}
#endif
#ifdef YYWILDCARD
{
int j = i - iLookAhead + YYWILDCARD;
if(
#if YY_SHIFT_MIN+YYWILDCARD<0
j>=0 &&
#endif
#if YY_SHIFT_MAX+YYWILDCARD>=YY_ACTTAB_COUNT
j<YY_ACTTAB_COUNT &&
#endif
yy_lookahead[j]==YYWILDCARD
){
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE, "%sWILDCARD %s => %s\n",
yyTracePrompt, yyTokenName[iLookAhead], yyTokenName[YYWILDCARD]);
}
#endif /* NDEBUG */
return yy_action[j];
}
}
#endif /* YYWILDCARD */
}
return yy_default[stateno];
}else{
return yy_action[i];
}
}
/*
** Find the appropriate action for a parser given the non-terminal
** look-ahead token iLookAhead.
**
** If the look-ahead token is YYNOCODE, then check to see if the action is
** independent of the look-ahead. If it is, return the action, otherwise
** return YY_NO_ACTION.
*/
static int yy_find_reduce_action(
int stateno, /* Current state number */
YYCODETYPE iLookAhead /* The look-ahead token */
){
int i;
#ifdef YYERRORSYMBOL
if( stateno>YY_REDUCE_COUNT ){
return yy_default[stateno];
}
#else
assert( stateno<=YY_REDUCE_COUNT );
#endif
i = yy_reduce_ofst[stateno];
assert( i!=YY_REDUCE_USE_DFLT );
assert( iLookAhead!=YYNOCODE );
i += iLookAhead;
#ifdef YYERRORSYMBOL
if( i<0 || i>=YY_ACTTAB_COUNT || yy_lookahead[i]!=iLookAhead ){
return yy_default[stateno];
}
#else
assert( i>=0 && i<YY_ACTTAB_COUNT );
assert( yy_lookahead[i]==iLookAhead );
#endif
return yy_action[i];
}
/*
** The following routine is called if the stack overflows.
*/
static void yyStackOverflow(yyParser *yypParser, YYMINORTYPE *yypMinor){
ParseARG_FETCH;
yypParser->yyidx--;
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE,"%sStack Overflow!\n",yyTracePrompt);
}
#endif
while( yypParser->yyidx>=0 ) yy_pop_parser_stack(yypParser);
/* Here code is inserted which will execute if the parser
** stack every overflows */
%%
ParseARG_STORE; /* Suppress warning about unused %extra_argument var */
}
/*
** Perform a shift action.
*/
static void yy_shift(
yyParser *yypParser, /* The parser to be shifted */
int yyNewState, /* The new state to shift in */
int yyMajor, /* The major token to shift in */
YYMINORTYPE *yypMinor /* Pointer to the minor token to shift in */
){
yyStackEntry *yytos;
yypParser->yyidx++;
#ifdef YYTRACKMAXSTACKDEPTH
if( yypParser->yyidx>yypParser->yyidxMax ){
yypParser->yyidxMax = yypParser->yyidx;
}
#endif
#if YYSTACKDEPTH>0
if( yypParser->yyidx>=YYSTACKDEPTH ){
yyStackOverflow(yypParser, yypMinor);
return;
}
#else
if( yypParser->yyidx>=yypParser->yystksz ){
yyGrowStack(yypParser);
if( yypParser->yyidx>=yypParser->yystksz ){
yyStackOverflow(yypParser, yypMinor);
return;
}
}
#endif
yytos = &yypParser->yystack[yypParser->yyidx];
yytos->stateno = (YYACTIONTYPE)yyNewState;
yytos->major = (YYCODETYPE)yyMajor;
yytos->minor = *yypMinor;
#ifndef NDEBUG
if( yyTraceFILE && yypParser->yyidx>0 ){
int i;
fprintf(yyTraceFILE,"%sShift %d\n",yyTracePrompt,yyNewState);
fprintf(yyTraceFILE,"%sStack:",yyTracePrompt);
for(i=1; i<=yypParser->yyidx; i++)
fprintf(yyTraceFILE," %s",yyTokenName[yypParser->yystack[i].major]);
fprintf(yyTraceFILE,"\n");
}
#endif
}
/* The following table contains information about every rule that
** is used during the reduce.
*/
static const struct {
YYCODETYPE lhs; /* Symbol on the left-hand side of the rule */
unsigned char nrhs; /* Number of right-hand side symbols in the rule */
} yyRuleInfo[] = {
%%
};
static void yy_accept(yyParser*); /* Forward Declaration */
/*
** Perform a reduce action and the shift that must immediately
** follow the reduce.
*/
static void yy_reduce(
yyParser *yypParser, /* The parser */
int yyruleno /* Number of the rule by which to reduce */
){
int yygoto; /* The next state */
int yyact; /* The next action */
YYMINORTYPE yygotominor; /* The LHS of the rule reduced */
yyStackEntry *yymsp; /* The top of the parser's stack */
int yysize; /* Amount to pop the stack */
ParseARG_FETCH;
yymsp = &yypParser->yystack[yypParser->yyidx];
#ifndef NDEBUG
if( yyTraceFILE && yyruleno>=0
&& yyruleno<(int)(sizeof(yyRuleName)/sizeof(yyRuleName[0])) ){
fprintf(yyTraceFILE, "%sReduce [%s].\n", yyTracePrompt,
yyRuleName[yyruleno]);
}
#endif /* NDEBUG */
/* Silence complaints from purify about yygotominor being uninitialized
** in some cases when it is copied into the stack after the following
** switch. yygotominor is uninitialized when a rule reduces that does
** not set the value of its left-hand side nonterminal. Leaving the
** value of the nonterminal uninitialized is utterly harmless as long
** as the value is never used. So really the only thing this code
** accomplishes is to quieten purify.
**
** 2007-01-16: The wireshark project (www.wireshark.org) reports that
** without this code, their parser segfaults. I'm not sure what there
** parser is doing to make this happen. This is the second bug report
** from wireshark this week. Clearly they are stressing Lemon in ways
** that it has not been previously stressed... (SQLite ticket #2172)
*/
/*memset(&yygotominor, 0, sizeof(yygotominor));*/
yygotominor = yyzerominor;
switch( yyruleno ){
/* Beginning here are the reduction cases. A typical example
** follows:
** case 0:
** #line <lineno> <grammarfile>
** { ... } // User supplied code
** #line <lineno> <thisfile>
** break;
*/
%%
};
yygoto = yyRuleInfo[yyruleno].lhs;
yysize = yyRuleInfo[yyruleno].nrhs;
yypParser->yyidx -= yysize;
yyact = yy_find_reduce_action(yymsp[-yysize].stateno,(YYCODETYPE)yygoto);
if( yyact < YYNSTATE ){
#ifdef NDEBUG
/* If we are not debugging and the reduce action popped at least
** one element off the stack, then we can push the new element back
** onto the stack here, and skip the stack overflow test in yy_shift().
** That gives a significant speed improvement. */
if( yysize ){
yypParser->yyidx++;
yymsp -= yysize-1;
yymsp->stateno = (YYACTIONTYPE)yyact;
yymsp->major = (YYCODETYPE)yygoto;
yymsp->minor = yygotominor;
}else
#endif
{
yy_shift(yypParser,yyact,yygoto,&yygotominor);
}
}else{
assert( yyact == YYNSTATE + YYNRULE + 1 );
yy_accept(yypParser);
}
}
/*
** The following code executes when the parse fails
*/
#ifndef YYNOERRORRECOVERY
static void yy_parse_failed(
yyParser *yypParser /* The parser */
){
ParseARG_FETCH;
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE,"%sFail!\n",yyTracePrompt);
}
#endif
while( yypParser->yyidx>=0 ) yy_pop_parser_stack(yypParser);
/* Here code is inserted which will be executed whenever the
** parser fails */
%%
ParseARG_STORE; /* Suppress warning about unused %extra_argument variable */
}
#endif /* YYNOERRORRECOVERY */
/*
** The following code executes when a syntax error first occurs.
*/
static void yy_syntax_error(
yyParser *yypParser, /* The parser */
int yymajor, /* The major type of the error token */
YYMINORTYPE yyminor /* The minor type of the error token */
){
ParseARG_FETCH;
#define TOKEN (yyminor.yy0)
%%
ParseARG_STORE; /* Suppress warning about unused %extra_argument variable */
}
/*
** The following is executed when the parser accepts
*/
static void yy_accept(
yyParser *yypParser /* The parser */
){
ParseARG_FETCH;
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE,"%sAccept!\n",yyTracePrompt);
}
#endif
while( yypParser->yyidx>=0 ) yy_pop_parser_stack(yypParser);
/* Here code is inserted which will be executed whenever the
** parser accepts */
%%
ParseARG_STORE; /* Suppress warning about unused %extra_argument variable */
}
/* The main parser program.
** The first argument is a pointer to a structure obtained from
** "ParseAlloc" which describes the current state of the parser.
** The second argument is the major token number. The third is
** the minor token. The fourth optional argument is whatever the
** user wants (and specified in the grammar) and is available for
** use by the action routines.
**
** Inputs:
** <ul>
** <li> A pointer to the parser (an opaque structure.)
** <li> The major token number.
** <li> The minor token number.
** <li> An option argument of a grammar-specified type.
** </ul>
**
** Outputs:
** None.
*/
void Parse(
void *yyp, /* The parser */
int yymajor, /* The major token code number */
ParseTOKENTYPE yyminor /* The value for the token */
ParseARG_PDECL /* Optional %extra_argument parameter */
){
YYMINORTYPE yyminorunion;
int yyact; /* The parser action. */
int yyendofinput; /* True if we are at the end of input */
#ifdef YYERRORSYMBOL
int yyerrorhit = 0; /* True if yymajor has invoked an error */
#endif
yyParser *yypParser; /* The parser */
/* (re)initialize the parser, if necessary */
yypParser = (yyParser*)yyp;
if( yypParser->yyidx<0 ){
#if YYSTACKDEPTH<=0
if( yypParser->yystksz <=0 ){
/*memset(&yyminorunion, 0, sizeof(yyminorunion));*/
yyminorunion = yyzerominor;
yyStackOverflow(yypParser, &yyminorunion);
return;
}
#endif
yypParser->yyidx = 0;
yypParser->yyerrcnt = -1;
yypParser->yystack[0].stateno = 0;
yypParser->yystack[0].major = 0;
}
yyminorunion.yy0 = yyminor;
yyendofinput = (yymajor==0);
ParseARG_STORE;
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE,"%sInput %s\n",yyTracePrompt,yyTokenName[yymajor]);
}
#endif
do{
yyact = yy_find_shift_action(yypParser,(YYCODETYPE)yymajor);
if( yyact<YYNSTATE ){
assert( !yyendofinput ); /* Impossible to shift the $ token */
yy_shift(yypParser,yyact,yymajor,&yyminorunion);
yypParser->yyerrcnt--;
yymajor = YYNOCODE;
}else if( yyact < YYNSTATE + YYNRULE ){
yy_reduce(yypParser,yyact-YYNSTATE);
}else{
assert( yyact == YY_ERROR_ACTION );
#ifdef YYERRORSYMBOL
int yymx;
#endif
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE,"%sSyntax Error!\n",yyTracePrompt);
}
#endif
#ifdef YYERRORSYMBOL
/* A syntax error has occurred.
** The response to an error depends upon whether or not the
** grammar defines an error token "ERROR".
**
** This is what we do if the grammar does define ERROR:
**
** * Call the %syntax_error function.
**
** * Begin popping the stack until we enter a state where
** it is legal to shift the error symbol, then shift
** the error symbol.
**
** * Set the error count to three.
**
** * Begin accepting and shifting new tokens. No new error
** processing will occur until three tokens have been
** shifted successfully.
**
*/
if( yypParser->yyerrcnt<0 ){
yy_syntax_error(yypParser,yymajor,yyminorunion);
}
yymx = yypParser->yystack[yypParser->yyidx].major;
if( yymx==YYERRORSYMBOL || yyerrorhit ){
#ifndef NDEBUG
if( yyTraceFILE ){
fprintf(yyTraceFILE,"%sDiscard input token %s\n",
yyTracePrompt,yyTokenName[yymajor]);
}
#endif
yy_destructor(yypParser, (YYCODETYPE)yymajor,&yyminorunion);
yymajor = YYNOCODE;
}else{
while(
yypParser->yyidx >= 0 &&
yymx != YYERRORSYMBOL &&
(yyact = yy_find_reduce_action(
yypParser->yystack[yypParser->yyidx].stateno,
YYERRORSYMBOL)) >= YYNSTATE
){
yy_pop_parser_stack(yypParser);
}
if( yypParser->yyidx < 0 || yymajor==0 ){
yy_destructor(yypParser,(YYCODETYPE)yymajor,&yyminorunion);
yy_parse_failed(yypParser);
yymajor = YYNOCODE;
}else if( yymx!=YYERRORSYMBOL ){
YYMINORTYPE u2;
u2.YYERRSYMDT = 0;
yy_shift(yypParser,yyact,YYERRORSYMBOL,&u2);
}
}
yypParser->yyerrcnt = 3;
yyerrorhit = 1;
#elif defined(YYNOERRORRECOVERY)
/* If the YYNOERRORRECOVERY macro is defined, then do not attempt to
** do any kind of error recovery. Instead, simply invoke the syntax
** error routine and continue going as if nothing had happened.
**
** Applications can set this macro (for example inside %include) if
** they intend to abandon the parse upon the first syntax error seen.
*/
yy_syntax_error(yypParser,yymajor,yyminorunion);
yy_destructor(yypParser,(YYCODETYPE)yymajor,&yyminorunion);
yymajor = YYNOCODE;
#else /* YYERRORSYMBOL is not defined */
/* This is what we do if the grammar does not define ERROR:
**
** * Report an error message, and throw away the input token.
**
** * If the input token is $, then fail the parse.
**
** As before, subsequent error messages are suppressed until
** three input tokens have been successfully shifted.
*/
if( yypParser->yyerrcnt<=0 ){
yy_syntax_error(yypParser,yymajor,yyminorunion);
}
yypParser->yyerrcnt = 3;
yy_destructor(yypParser,(YYCODETYPE)yymajor,&yyminorunion);
if( yyendofinput ){
yy_parse_failed(yypParser);
}
yymajor = YYNOCODE;
#endif
}
}while( yymajor!=YYNOCODE && yypParser->yyidx>=0 );
return;
}

View file

@ -0,0 +1,43 @@
%name pluralParse
%token_type {int}
%extra_argument {struct parse_state *s}
%right T_QMARK.
%left T_OR.
%left T_AND.
%left T_EQ T_NE.
%left T_LT T_LE T_GT T_GE.
%left T_ADD T_SUB.
%left T_MUL T_DIV T_MOD.
%right T_NOT.
%nonassoc T_COLON T_N T_LPAREN T_RPAREN.
%include {
#include <assert.h>
struct parse_state {
int num;
int res;
};
}
input ::= expr(A). { s->res = A; }
expr(A) ::= expr(B) T_QMARK expr(C) T_COLON expr(D). { A = B ? C : D; }
expr(A) ::= expr(B) T_OR expr(C). { A = B || C; }
expr(A) ::= expr(B) T_AND expr(C). { A = B && C; }
expr(A) ::= expr(B) T_EQ expr(C). { A = B == C; }
expr(A) ::= expr(B) T_NE expr(C). { A = B != C; }
expr(A) ::= expr(B) T_LT expr(C). { A = B < C; }
expr(A) ::= expr(B) T_LE expr(C). { A = B <= C; }
expr(A) ::= expr(B) T_GT expr(C). { A = B > C; }
expr(A) ::= expr(B) T_GE expr(C). { A = B >= C; }
expr(A) ::= expr(B) T_ADD expr(C). { A = B + C; }
expr(A) ::= expr(B) T_SUB expr(C). { A = B - C; }
expr(A) ::= expr(B) T_MUL expr(C). { A = B * C; }
expr(A) ::= expr(B) T_DIV expr(C). { A = B / C; }
expr(A) ::= expr(B) T_MOD expr(C). { A = B % C; }
expr(A) ::= T_NOT expr(B). { A = !B; }
expr(A) ::= T_N. { A = s->num; }
expr(A) ::= T_NUM(B). { A = B; }
expr(A) ::= T_LPAREN expr(B) T_RPAREN. { A = B; }

View file

@ -17,6 +17,7 @@
*/
#include "template_lmo.h"
#include "plural_formula.h"
/*
* Hash function from http://www.azillionmonkeys.com/qed/hash.html
@ -69,17 +70,51 @@ uint32_t sfh_hash(const char *data, int len)
return hash;
}
uint32_t lmo_canon_hash(const char *str, int len)
uint32_t lmo_canon_hash(const char *str, int len,
const char *ctx, int ctxlen, int plural)
{
char res[4096];
char *ptr, prev;
char *ptr, *end, prev;
int off;
if (!str || len >= sizeof(res))
if (!str)
return 0;
for (prev = ' ', ptr = res, off = 0; off < len; prev = *str, off++, str++)
ptr = res;
end = res + sizeof(res);
if (ctx)
{
for (prev = ' ', off = 0; off < ctxlen; prev = *ctx, off++, ctx++)
{
if (ptr >= end)
return 0;
if (isspace(*ctx))
{
if (!isspace(prev))
*ptr++ = ' ';
}
else
{
*ptr++ = *ctx;
}
}
if ((ptr > res) && isspace(*(ptr-1)))
ptr--;
if (ptr >= end)
return 0;
*ptr++ = '\1';
}
for (prev = ' ', off = 0; off < len; prev = *str, off++, str++)
{
if (ptr >= end)
return 0;
if (isspace(*str))
{
if (!isspace(prev))
@ -94,6 +129,14 @@ uint32_t lmo_canon_hash(const char *str, int len)
if ((ptr > res) && isspace(*(ptr-1)))
ptr--;
if (plural > -1)
{
if (plural >= 100 || ptr + 3 >= end)
return 0;
ptr += snprintf(ptr, 3, "\2%d", plural);
}
return sfh_hash(res, ptr - res);
}
@ -277,7 +320,193 @@ static lmo_entry_t * lmo_find_entry(lmo_archive_t *ar, uint32_t hash)
return NULL;
}
void *pluralParseAlloc(void *(*)(size_t));
void pluralParse(void *, int, int, void *);
void pluralParseFree(void *, void (*)(void *));
static int lmo_eval_plural(const char *expr, int len, int val)
{
struct { int num; int res; } s = { .num = val, .res = -1 };
const char *p = NULL;
void *pParser = NULL;
int t, n;
char c;
while (len > 7) {
if (*expr == 'p') {
if (!strncmp(expr, "plural=", 7)) {
p = expr + 7;
len -= 7;
break;
}
}
expr++;
len--;
}
if (!p)
goto out;
pParser = pluralParseAlloc(malloc);
if (!pParser)
goto out;
while (len-- > 0) {
c = *p++;
t = -1;
n = 0;
switch (c) {
case ' ':
case '\t':
continue;
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
t = T_NUM;
n = c - '0';
while (*p >= '0' && *p <= '9') {
n *= 10;
n += *p - '0';
p++;
}
break;
case '=':
if (*p == '=') {
t = T_EQ;
p++;
}
break;
case '!':
if (*p == '=') {
t = T_NE;
p++;
}
else {
t = T_NOT;
}
break;
case '&':
if (*p == '&') {
t = T_AND;
p++;
}
break;
case '|':
if (*p == '|') {
t = T_OR;
p++;
}
break;
case '<':
if (*p == '=') {
t = T_LE;
p++;
}
else {
t = T_LT;
}
break;
case '>':
if (*p == '=') {
t = T_GE;
p++;
}
else {
t = T_GT;
}
break;
case '*':
t = T_MUL;
break;
case '/':
t = T_DIV;
break;
case '%':
t = T_MOD;
break;
case '+':
t = T_ADD;
break;
case '-':
t = T_SUB;
break;
case 'n':
t = T_N;
break;
case '?':
t = T_QMARK;
break;
case ':':
t = T_COLON;
break;
case '(':
t = T_LPAREN;
break;
case ')':
t = T_RPAREN;
break;
case ';':
case '\n':
case '\0':
t = 0;
break;
}
/* syntax error */
if (t < 0)
goto out;
pluralParse(pParser, t, n, &s);
/* eof */
if (t == 0)
break;
}
pluralParse(pParser, 0, 0, &s);
out:
pluralParseFree(pParser, free);
return s.res;
}
int lmo_translate(const char *key, int keylen, char **out, int *outlen)
{
return lmo_translate_ctxt(key, keylen, NULL, 0, out, outlen);
}
int lmo_translate_ctxt(const char *key, int keylen,
const char *ctx, int ctxlen,
char **out, int *outlen)
{
uint32_t hash;
lmo_entry_t *e;
@ -286,7 +515,62 @@ int lmo_translate(const char *key, int keylen, char **out, int *outlen)
if (!key || !_lmo_active_catalog)
return -2;
hash = lmo_canon_hash(key, keylen);
hash = lmo_canon_hash(key, keylen, ctx, ctxlen, -1);
if (hash > 0)
{
for (ar = _lmo_active_catalog->archives; ar; ar = ar->next)
{
if ((e = lmo_find_entry(ar, hash)) != NULL)
{
*out = ar->mmap + ntohl(e->offset);
*outlen = ntohl(e->length);
return 0;
}
}
}
return -1;
}
int lmo_translate_plural(int n, const char *skey, int skeylen,
const char *pkey, int pkeylen,
char **out, int *outlen)
{
return lmo_translate_plural_ctxt(n, skey, skeylen, pkey, pkeylen,
NULL, 0, out, outlen);
}
int lmo_translate_plural_ctxt(int n, const char *skey, int skeylen,
const char *pkey, int pkeylen,
const char *ctx, int ctxlen,
char **out, int *outlen)
{
int pid = -1;
uint32_t hash;
lmo_entry_t *e;
lmo_archive_t *ar;
const char *plural_formula;
if (!skey || !pkey || !_lmo_active_catalog)
return -2;
for (ar = _lmo_active_catalog->archives; ar; ar = ar->next) {
e = lmo_find_entry(ar, 0);
if (e != NULL) {
pid = lmo_eval_plural(ar->mmap + ntohl(e->offset), ntohl(e->length), n);
break;
}
}
if (pid == -1)
pid = (n != 1);
hash = lmo_canon_hash(skey, skeylen, ctx, ctxlen, pid);
if (hash == 0)
return -1;
for (ar = _lmo_active_catalog->archives; ar; ar = ar->next)
{
@ -298,7 +582,18 @@ int lmo_translate(const char *key, int keylen, char **out, int *outlen)
}
}
return -1;
if (n != 1)
{
*out = (char *)pkey;
*outlen = pkeylen;
}
else
{
*out = (char *)skey;
*outlen = skeylen;
}
return 0;
}
void lmo_iterate(lmo_iterate_cb_t cb, void *priv)

View file

@ -76,7 +76,8 @@ typedef struct lmo_catalog lmo_catalog_t;
typedef void (*lmo_iterate_cb_t)(uint32_t, const char *, int, void *);
uint32_t sfh_hash(const char *data, int len);
uint32_t lmo_canon_hash(const char *data, int len);
uint32_t lmo_canon_hash(const char *data, int len,
const char *ctx, int ctxlen, int plural);
lmo_archive_t * lmo_open(const char *file);
void lmo_close(lmo_archive_t *ar);
@ -88,6 +89,15 @@ extern lmo_catalog_t *_lmo_active_catalog;
int lmo_load_catalog(const char *lang, const char *dir);
int lmo_change_catalog(const char *lang);
int lmo_translate(const char *key, int keylen, char **out, int *outlen);
int lmo_translate_ctxt(const char *key, int keylen,
const char *ctx, int ctxlen, char **out, int *outlen);
int lmo_translate_plural(int n, const char *skey, int skeylen,
const char *pkey, int pkeylen,
char **out, int *outlen);
int lmo_translate_plural_ctxt(int n, const char *skey, int skeylen,
const char *pkey, int pkeylen,
const char *ctx, int ctxlen,
char **out, int *outlen);
void lmo_iterate(lmo_iterate_cb_t cb, void *priv);
void lmo_close_catalog(const char *lang);

View file

@ -148,12 +148,37 @@ static int template_L_get_translations(lua_State *L) {
}
static int template_L_translate(lua_State *L) {
size_t len;
size_t len, ctxlen = 0;
char *tr;
int trlen;
const char *key = luaL_checklstring(L, 1, &len);
const char *ctx = luaL_optlstring(L, 2, NULL, &ctxlen);
switch (lmo_translate(key, len, &tr, &trlen))
switch (lmo_translate_ctxt(key, len, ctx, ctxlen, &tr, &trlen))
{
case 0:
lua_pushlstring(L, tr, trlen);
return 1;
case -1:
return 0;
}
lua_pushnil(L);
lua_pushstring(L, "no catalog loaded");
return 2;
}
static int template_L_ntranslate(lua_State *L) {
size_t slen, plen, ctxlen = 0;
char *tr;
int trlen;
int n = luaL_checkinteger(L, 1);
const char *skey = luaL_checklstring(L, 2, &slen);
const char *pkey = luaL_checklstring(L, 3, &plen);
const char *ctx = luaL_optlstring(L, 4, NULL, &ctxlen);
switch (lmo_translate_plural_ctxt(n, skey, slen, pkey, plen, ctx, ctxlen, &tr, &trlen))
{
case 0:
lua_pushlstring(L, tr, trlen);
@ -188,6 +213,7 @@ static const luaL_reg R[] = {
{ "change_catalog", template_L_change_catalog },
{ "get_translations", template_L_get_translations },
{ "translate", template_L_translate },
{ "ntranslate", template_L_ntranslate },
{ "hash", template_L_hash },
{ NULL, NULL }
};

View file

@ -474,10 +474,26 @@ void luastr_escape(struct template_buffer *out, const char *s, unsigned int l,
void luastr_translate(struct template_buffer *out, const char *s, unsigned int l,
int escape_xml)
{
int trlen, idlen = l, ctxtlen = 0, esc = 0;
const char *p, *msgid = s, *msgctxt = NULL;
char *tr;
int trlen;
if (!lmo_translate(s, l, &tr, &trlen))
for (p = s; p < s + l; p++) {
if (esc) {
esc = 0;
}
else if (*p == '\\') {
esc = 1;
}
else if (*p == '|') {
idlen = p - s;
msgctxt = p + 1;
ctxtlen = s + l - msgctxt;
break;
}
}
if (!lmo_translate_ctxt(msgid, idlen, msgctxt, ctxtlen, &tr, &trlen))
luastr_escape(out, tr, trlen, escape_xml);
else
luastr_escape(out, s, l, escape_xml);