Add Mde String and PrintLibs. Port StrGather to the Mde Unicode implementation. Compi...
[mirror_edk2.git] / Tools / Source / TianoTools / StrGather / StrGather.c
1 /*++
2
3 Copyright (c) 2004, Intel Corporation
4 All rights reserved. This program and the accompanying materials
5 are licensed and made available under the terms and conditions of the BSD License
6 which accompanies this distribution. The full text of the license may be found at
7 http://opensource.org/licenses/bsd-license.php
8
9 THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,
10 WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
11
12 Module Name:
13
14 StrGather.c
15
16 Abstract:
17
18 Parse a strings file and create or add to a string database file.
19
20 --*/
21
22 #include <stdio.h>
23 #include <string.h>
24 #include <stdlib.h>
25 #include <ctype.h>
26
27 #include <Base.h>
28 #include <UefiBaseTypes.h>
29 #include <CommonLib.h>
30 #include "EfiUtilityMsgs.h"
31 #include "StrGather.h"
32 #include "StringDB.h"
33
34 #define TOOL_VERSION "0.31"
35
36 #ifndef MAX_PATH
37 #define MAX_PATH 255
38 #endif
39 #define MAX_NEST_DEPTH 20 // just in case we get in an endless loop.
40 #define MAX_STRING_IDENTIFIER_NAME 100 // number of wchars
41 #define MAX_LINE_LEN 200
42 #define STRING_TOKEN "STRING_TOKEN"
43 #define DEFAULT_BASE_NAME "BaseName"
44 //
45 // Operational modes for this utility
46 //
47 #define MODE_UNKNOWN 0
48 #define MODE_PARSE 1
49 #define MODE_SCAN 2
50 #define MODE_DUMP 3
51
52 //
53 // We keep a linked list of these for the source files we process
54 //
55 typedef struct _SOURCE_FILE {
56 FILE *Fptr;
57 WCHAR *FileBuffer;
58 WCHAR *FileBufferPtr;
59 UINT32 FileSize;
60 INT8 FileName[MAX_PATH];
61 UINT32 LineNum;
62 BOOLEAN EndOfFile;
63 BOOLEAN SkipToHash;
64 struct _SOURCE_FILE *Previous;
65 struct _SOURCE_FILE *Next;
66 WCHAR ControlCharacter;
67 } SOURCE_FILE;
68
69 #define DEFAULT_CONTROL_CHARACTER UNICODE_SLASH
70
71 //
72 // Here's all our globals. We need a linked list of include paths, a linked
73 // list of source files, a linked list of subdirectories (appended to each
74 // include path when searching), and a couple other fields.
75 //
76 static struct {
77 SOURCE_FILE SourceFiles;
78 TEXT_STRING_LIST *IncludePaths; // all include paths to search
79 TEXT_STRING_LIST *LastIncludePath;
80 TEXT_STRING_LIST *ScanFileName;
81 TEXT_STRING_LIST *LastScanFileName;
82 TEXT_STRING_LIST *SkipExt; // if -skipext .uni
83 TEXT_STRING_LIST *LastSkipExt;
84 TEXT_STRING_LIST *IndirectionFileName;
85 TEXT_STRING_LIST *LastIndirectionFileName;
86 TEXT_STRING_LIST *DatabaseFileName;
87 TEXT_STRING_LIST *LastDatabaseFileName;
88 WCHAR_STRING_LIST *Language;
89 WCHAR_STRING_LIST *LastLanguage;
90 WCHAR_MATCHING_STRING_LIST *IndirectionList; // from indirection file(s)
91 WCHAR_MATCHING_STRING_LIST *LastIndirectionList;
92 BOOLEAN Verbose; // for more detailed output
93 BOOLEAN VerboseDatabaseWrite; // for more detailed output when writing database
94 BOOLEAN VerboseDatabaseRead; // for more detailed output when reading database
95 BOOLEAN NewDatabase; // to start from scratch
96 BOOLEAN IgnoreNotFound; // when scanning
97 BOOLEAN VerboseScan;
98 BOOLEAN UnquotedStrings; // -uqs option
99 INT8 OutputDatabaseFileName[MAX_PATH];
100 INT8 StringHFileName[MAX_PATH];
101 INT8 StringCFileName[MAX_PATH]; // output .C filename
102 INT8 DumpUFileName[MAX_PATH]; // output unicode dump file name
103 INT8 HiiExportPackFileName[MAX_PATH]; // HII export pack file name
104 INT8 BaseName[MAX_PATH]; // base filename of the strings file
105 UINT32 Mode;
106 } mGlobals;
107
108 static
109 BOOLEAN
110 IsValidIdentifierChar (
111 INT8 Char,
112 BOOLEAN FirstChar
113 );
114
115 static
116 void
117 RewindFile (
118 SOURCE_FILE *SourceFile
119 );
120
121 static
122 BOOLEAN
123 SkipTo (
124 SOURCE_FILE *SourceFile,
125 WCHAR WChar,
126 BOOLEAN StopAfterNewline
127 );
128
129 static
130 UINT32
131 SkipWhiteSpace (
132 SOURCE_FILE *SourceFile
133 );
134
135 static
136 BOOLEAN
137 IsWhiteSpace (
138 SOURCE_FILE *SourceFile
139 );
140
141 static
142 BOOLEAN
143 EndOfFile (
144 SOURCE_FILE *SourceFile
145 );
146
147 static
148 void
149 PreprocessFile (
150 SOURCE_FILE *SourceFile
151 );
152
153 static
154 UINT32
155 GetStringIdentifierName (
156 IN SOURCE_FILE *SourceFile,
157 IN OUT WCHAR *StringIdentifierName,
158 IN UINT32 StringIdentifierNameLen
159 );
160
161 static
162 UINT32
163 GetLanguageIdentifierName (
164 IN SOURCE_FILE *SourceFile,
165 IN OUT WCHAR *LanguageIdentifierName,
166 IN UINT32 LanguageIdentifierNameLen,
167 IN BOOLEAN Optional
168 );
169
170 static
171 WCHAR *
172 GetPrintableLanguageName (
173 IN SOURCE_FILE *SourceFile
174 );
175
176 static
177 STATUS
178 AddCommandLineLanguage (
179 IN INT8 *Language
180 );
181
182 static
183 WCHAR *
184 GetQuotedString (
185 SOURCE_FILE *SourceFile,
186 BOOLEAN Optional
187 );
188
189 static
190 STATUS
191 ProcessIncludeFile (
192 SOURCE_FILE *SourceFile,
193 SOURCE_FILE *ParentSourceFile
194 );
195
196 static
197 STATUS
198 ParseFile (
199 SOURCE_FILE *SourceFile
200 );
201
202 static
203 FILE *
204 FindFile (
205 IN INT8 *FileName,
206 OUT INT8 *FoundFileName,
207 IN UINT32 FoundFileNameLen
208 );
209
210 static
211 STATUS
212 ProcessArgs (
213 int Argc,
214 char *Argv[]
215 );
216
217 static
218 STATUS
219 ProcessFile (
220 SOURCE_FILE *SourceFile
221 );
222
223 static
224 UINT32
225 wstrcmp (
226 WCHAR *Buffer,
227 WCHAR *Str
228 );
229
230 static
231 void
232 Usage (
233 VOID
234 );
235
236 static
237 void
238 FreeLists (
239 VOID
240 );
241
242 static
243 void
244 ProcessTokenString (
245 SOURCE_FILE *SourceFile
246 );
247
248 static
249 void
250 ProcessTokenInclude (
251 SOURCE_FILE *SourceFile
252 );
253
254 static
255 void
256 ProcessTokenScope (
257 SOURCE_FILE *SourceFile
258 );
259
260 static
261 void
262 ProcessTokenLanguage (
263 SOURCE_FILE *SourceFile
264 );
265
266 static
267 void
268 ProcessTokenLangDef (
269 SOURCE_FILE *SourceFile
270 );
271
272 static
273 STATUS
274 ScanFiles (
275 TEXT_STRING_LIST *ScanFiles
276 );
277
278 static
279 STATUS
280 ParseIndirectionFiles (
281 TEXT_STRING_LIST *Files
282 );
283
284 STATUS
285 StringDBCreateHiiExportPack (
286 INT8 *OutputFileName
287 );
288
289 int
290 main (
291 int Argc,
292 char *Argv[]
293 )
294 /*++
295
296 Routine Description:
297
298 Call the routine to parse the command-line options, then process the file.
299
300 Arguments:
301
302 Argc - Standard C main() argc and argv.
303 Argv - Standard C main() argc and argv.
304
305 Returns:
306
307 0 if successful
308 nonzero otherwise
309
310 --*/
311 {
312 STATUS Status;
313
314 SetUtilityName (PROGRAM_NAME);
315 //
316 // Process the command-line arguments
317 //
318 Status = ProcessArgs (Argc, Argv);
319 if (Status != STATUS_SUCCESS) {
320 return Status;
321 }
322 //
323 // Initialize the database manager
324 //
325 StringDBConstructor ();
326 //
327 // We always try to read in an existing database file. It may not
328 // exist, which is ok usually.
329 //
330 if (mGlobals.NewDatabase == 0) {
331 //
332 // Read all databases specified.
333 //
334 for (mGlobals.LastDatabaseFileName = mGlobals.DatabaseFileName;
335 mGlobals.LastDatabaseFileName != NULL;
336 mGlobals.LastDatabaseFileName = mGlobals.LastDatabaseFileName->Next
337 ) {
338 Status = StringDBReadDatabase (mGlobals.LastDatabaseFileName->Str, TRUE, mGlobals.VerboseDatabaseRead);
339 if (Status != STATUS_SUCCESS) {
340 return Status;
341 }
342 }
343 }
344 //
345 // Read indirection file(s) if specified
346 //
347 if (ParseIndirectionFiles (mGlobals.IndirectionFileName) != STATUS_SUCCESS) {
348 goto Finish;
349 }
350 //
351 // If scanning source files, do that now
352 //
353 if (mGlobals.Mode == MODE_SCAN) {
354 ScanFiles (mGlobals.ScanFileName);
355 } else if (mGlobals.Mode == MODE_PARSE) {
356 //
357 // Parsing a unicode strings file
358 //
359 mGlobals.SourceFiles.ControlCharacter = DEFAULT_CONTROL_CHARACTER;
360 Status = ProcessIncludeFile (&mGlobals.SourceFiles, NULL);
361 if (Status != STATUS_SUCCESS) {
362 goto Finish;
363 }
364 }
365 //
366 // Create the string defines header file if there have been no errors.
367 //
368 ParserSetPosition (NULL, 0);
369 if ((mGlobals.StringHFileName[0] != 0) && (GetUtilityStatus () < STATUS_ERROR)) {
370 Status = StringDBDumpStringDefines (mGlobals.StringHFileName, mGlobals.BaseName);
371 if (Status != EFI_SUCCESS) {
372 goto Finish;
373 }
374 }
375 //
376 // Dump the strings to a .c file if there have still been no errors.
377 //
378 if ((mGlobals.StringCFileName[0] != 0) && (GetUtilityStatus () < STATUS_ERROR)) {
379 Status = StringDBDumpCStrings (
380 mGlobals.StringCFileName,
381 mGlobals.BaseName,
382 mGlobals.Language,
383 mGlobals.IndirectionList
384 );
385 if (Status != EFI_SUCCESS) {
386 goto Finish;
387 }
388 }
389 //
390 // Dump the database if requested
391 //
392 if ((mGlobals.DumpUFileName[0] != 0) && (GetUtilityStatus () < STATUS_ERROR)) {
393 StringDBDumpDatabase (NULL, mGlobals.DumpUFileName, FALSE);
394 }
395 //
396 // Dump the string data as HII binary string pack if requested
397 //
398 if ((mGlobals.HiiExportPackFileName[0] != 0) && (GetUtilityStatus () < STATUS_ERROR)) {
399 StringDBCreateHiiExportPack (mGlobals.HiiExportPackFileName);
400 }
401 //
402 // Always update the database if no errors and not in dump mode. If they specified -od
403 // for an output database file name, then use that name. Otherwise use the name of
404 // the first database file specified with -db
405 //
406 if ((mGlobals.Mode != MODE_DUMP) && (GetUtilityStatus () < STATUS_ERROR)) {
407 if (mGlobals.OutputDatabaseFileName[0]) {
408 Status = StringDBWriteDatabase (mGlobals.OutputDatabaseFileName, mGlobals.VerboseDatabaseWrite);
409 } else {
410 Status = StringDBWriteDatabase (mGlobals.DatabaseFileName->Str, mGlobals.VerboseDatabaseWrite);
411 }
412
413 if (Status != EFI_SUCCESS) {
414 goto Finish;
415 }
416 }
417
418 Finish:
419 //
420 // Free up memory
421 //
422 FreeLists ();
423 StringDBDestructor ();
424 return GetUtilityStatus ();
425 }
426
427 static
428 STATUS
429 ProcessIncludeFile (
430 SOURCE_FILE *SourceFile,
431 SOURCE_FILE *ParentSourceFile
432 )
433 /*++
434
435 Routine Description:
436
437 Given a source file, open the file and parse it
438
439 Arguments:
440
441 SourceFile - name of file to parse
442 ParentSourceFile - for error reporting purposes, the file that #included SourceFile.
443
444 Returns:
445
446 Standard status.
447
448 --*/
449 {
450 static UINT32 NestDepth = 0;
451 INT8 FoundFileName[MAX_PATH];
452 STATUS Status;
453
454 Status = STATUS_SUCCESS;
455 NestDepth++;
456 //
457 // Print the file being processed. Indent so you can tell the include nesting
458 // depth.
459 //
460 if (mGlobals.Verbose) {
461 fprintf (stdout, "%*cProcessing file '%s'\n", NestDepth * 2, ' ', SourceFile->FileName);
462 }
463
464 //
465 // Make sure we didn't exceed our maximum nesting depth
466 //
467 if (NestDepth > MAX_NEST_DEPTH) {
468 Error (NULL, 0, 0, SourceFile->FileName, "max nesting depth (%d) exceeded", NestDepth);
469 Status = STATUS_ERROR;
470 goto Finish;
471 }
472 //
473 // Try to open the file locally, and if that fails try along our include paths.
474 //
475 strcpy (FoundFileName, SourceFile->FileName);
476 if ((SourceFile->Fptr = fopen (FoundFileName, "rb")) == NULL) {
477 //
478 // Try to find it among the paths if it has a parent (that is, it is included
479 // by someone else).
480 //
481 if (ParentSourceFile == NULL) {
482 Error (NULL, 0, 0, SourceFile->FileName, "file not found");
483 return STATUS_ERROR;
484 }
485
486 SourceFile->Fptr = FindFile (SourceFile->FileName, FoundFileName, sizeof (FoundFileName));
487 if (SourceFile->Fptr == NULL) {
488 Error (ParentSourceFile->FileName, ParentSourceFile->LineNum, 0, SourceFile->FileName, "include file not found");
489 return STATUS_ERROR;
490 }
491 }
492 //
493 // Process the file found
494 //
495 ProcessFile (SourceFile);
496 Finish:
497 //
498 // Close open files and return status
499 //
500 if (SourceFile->Fptr != NULL) {
501 fclose (SourceFile->Fptr);
502 }
503
504 return Status;
505 }
506
507 static
508 STATUS
509 ProcessFile (
510 SOURCE_FILE *SourceFile
511 )
512 {
513 //
514 // Get the file size, and then read the entire thing into memory.
515 // Allocate space for a terminator character.
516 //
517 fseek (SourceFile->Fptr, 0, SEEK_END);
518 SourceFile->FileSize = ftell (SourceFile->Fptr);
519 fseek (SourceFile->Fptr, 0, SEEK_SET);
520 SourceFile->FileBuffer = (WCHAR *) malloc (SourceFile->FileSize + sizeof (WCHAR));
521 if (SourceFile->FileBuffer == NULL) {
522 Error (NULL, 0, 0, "memory allocation failure", NULL);
523 return STATUS_ERROR;
524 }
525
526 fread ((VOID *) SourceFile->FileBuffer, SourceFile->FileSize, 1, SourceFile->Fptr);
527 SourceFile->FileBuffer[(SourceFile->FileSize / sizeof (WCHAR))] = UNICODE_NULL;
528 //
529 // Pre-process the file to replace comments with spaces
530 //
531 PreprocessFile (SourceFile);
532 //
533 // Parse the file
534 //
535 ParseFile (SourceFile);
536 free (SourceFile->FileBuffer);
537 return STATUS_SUCCESS;
538 }
539
540 static
541 STATUS
542 ParseFile (
543 SOURCE_FILE *SourceFile
544 )
545 {
546 BOOLEAN InComment;
547 UINT32 Len;
548
549 //
550 // First character of a unicode file is special. Make sure
551 //
552 if (SourceFile->FileBufferPtr[0] != UNICODE_FILE_START) {
553 Error (SourceFile->FileName, 1, 0, SourceFile->FileName, "file does not appear to be a unicode file");
554 return STATUS_ERROR;
555 }
556
557 SourceFile->FileBufferPtr++;
558 InComment = FALSE;
559 //
560 // Print the first line if in verbose mode
561 //
562 if (mGlobals.Verbose) {
563 printf ("%d: %S\n", SourceFile->LineNum, SourceFile->FileBufferPtr);
564 }
565 //
566 // Since the syntax is relatively straightforward, just switch on the next char
567 //
568 while (!EndOfFile (SourceFile)) {
569 //
570 // Check for whitespace
571 //
572 if (SourceFile->FileBufferPtr[0] == UNICODE_SPACE) {
573 SourceFile->FileBufferPtr++;
574 } else if (SourceFile->FileBufferPtr[0] == UNICODE_TAB) {
575 SourceFile->FileBufferPtr++;
576 } else if (SourceFile->FileBufferPtr[0] == UNICODE_CR) {
577 SourceFile->FileBufferPtr++;
578 } else if (SourceFile->FileBufferPtr[0] == UNICODE_LF) {
579 SourceFile->FileBufferPtr++;
580 SourceFile->LineNum++;
581 if (mGlobals.Verbose) {
582 printf ("%d: %S\n", SourceFile->LineNum, SourceFile->FileBufferPtr);
583 }
584
585 InComment = FALSE;
586 } else if (SourceFile->FileBufferPtr[0] == 0) {
587 SourceFile->FileBufferPtr++;
588 } else if (InComment) {
589 SourceFile->FileBufferPtr++;
590 } else if ((SourceFile->FileBufferPtr[0] == UNICODE_SLASH) && (SourceFile->FileBufferPtr[1] == UNICODE_SLASH)) {
591 SourceFile->FileBufferPtr += 2;
592 InComment = TRUE;
593 } else if (SourceFile->SkipToHash && (SourceFile->FileBufferPtr[0] != SourceFile->ControlCharacter)) {
594 SourceFile->FileBufferPtr++;
595 } else {
596 SourceFile->SkipToHash = FALSE;
597 if ((SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter) &&
598 ((Len = wstrcmp (SourceFile->FileBufferPtr + 1, L"include")) > 0)
599 ) {
600 SourceFile->FileBufferPtr += Len + 1;
601 ProcessTokenInclude (SourceFile);
602 } else if ((SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter) &&
603 (Len = wstrcmp (SourceFile->FileBufferPtr + 1, L"scope")) > 0
604 ) {
605 SourceFile->FileBufferPtr += Len + 1;
606 ProcessTokenScope (SourceFile);
607 } else if ((SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter) &&
608 (Len = wstrcmp (SourceFile->FileBufferPtr + 1, L"language")) > 0
609 ) {
610 SourceFile->FileBufferPtr += Len + 1;
611 ProcessTokenLanguage (SourceFile);
612 } else if ((SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter) &&
613 (Len = wstrcmp (SourceFile->FileBufferPtr + 1, L"langdef")) > 0
614 ) {
615 SourceFile->FileBufferPtr += Len + 1;
616 ProcessTokenLangDef (SourceFile);
617 } else if ((SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter) &&
618 (Len = wstrcmp (SourceFile->FileBufferPtr + 1, L"string")) > 0
619 ) {
620 SourceFile->FileBufferPtr += Len + 1;
621 ProcessTokenString (SourceFile);
622 } else if ((SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter) &&
623 (Len = wstrcmp (SourceFile->FileBufferPtr + 1, L"EFI_BREAKPOINT()")) > 0
624 ) {
625 SourceFile->FileBufferPtr += Len;
626 //
627 // BUGBUG: Caling EFI_BREAKOINT() is breaking the link. What is the proper action for this tool
628 // in this condition?
629 //
630 // EFI_BREAKPOINT ();
631 } else if ((SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter) &&
632 (SourceFile->FileBufferPtr[1] == UNICODE_EQUAL_SIGN)
633 ) {
634 SourceFile->ControlCharacter = SourceFile->FileBufferPtr[2];
635 SourceFile->FileBufferPtr += 3;
636 } else {
637 Error (SourceFile->FileName, SourceFile->LineNum, 0, "unrecognized token", "%S", SourceFile->FileBufferPtr);
638 //
639 // Treat rest of line as a comment.
640 //
641 InComment = TRUE;
642 }
643 }
644 }
645
646 return STATUS_SUCCESS;
647 }
648
649 static
650 void
651 PreprocessFile (
652 SOURCE_FILE *SourceFile
653 )
654 /*++
655
656 Routine Description:
657 Preprocess a file to replace all carriage returns with NULLs so
658 we can print lines from the file to the screen.
659
660 Arguments:
661 SourceFile - structure that we use to keep track of an input file.
662
663 Returns:
664 Nothing.
665
666 --*/
667 {
668 BOOLEAN InComment;
669
670 RewindFile (SourceFile);
671 InComment = FALSE;
672 while (!EndOfFile (SourceFile)) {
673 //
674 // If a line-feed, then no longer in a comment
675 //
676 if (SourceFile->FileBufferPtr[0] == UNICODE_LF) {
677 SourceFile->FileBufferPtr++;
678 SourceFile->LineNum++;
679 InComment = 0;
680 } else if (SourceFile->FileBufferPtr[0] == UNICODE_CR) {
681 //
682 // Replace all carriage returns with a NULL so we can print stuff
683 //
684 SourceFile->FileBufferPtr[0] = 0;
685 SourceFile->FileBufferPtr++;
686 } else if (InComment) {
687 SourceFile->FileBufferPtr[0] = UNICODE_SPACE;
688 SourceFile->FileBufferPtr++;
689 } else if ((SourceFile->FileBufferPtr[0] == UNICODE_SLASH) && (SourceFile->FileBufferPtr[1] == UNICODE_SLASH)) {
690 SourceFile->FileBufferPtr += 2;
691 InComment = TRUE;
692 } else {
693 SourceFile->FileBufferPtr++;
694 }
695 }
696 //
697 // Could check for end-of-file and still in a comment, but
698 // should not be necessary. So just restore the file pointers.
699 //
700 RewindFile (SourceFile);
701 }
702
703 static
704 WCHAR *
705 GetPrintableLanguageName (
706 IN SOURCE_FILE *SourceFile
707 )
708 {
709 WCHAR *String;
710 WCHAR *Start;
711 WCHAR *Ptr;
712 UINT32 Len;
713
714 SkipWhiteSpace (SourceFile);
715 if (SourceFile->FileBufferPtr[0] != UNICODE_DOUBLE_QUOTE) {
716 Error (
717 SourceFile->FileName,
718 SourceFile->LineNum,
719 0,
720 "expected quoted printable language name",
721 "%S",
722 SourceFile->FileBufferPtr
723 );
724 SourceFile->SkipToHash = TRUE;
725 return NULL;
726 }
727
728 Len = 0;
729 SourceFile->FileBufferPtr++;
730 Start = Ptr = SourceFile->FileBufferPtr;
731 while (!EndOfFile (SourceFile)) {
732 if (SourceFile->FileBufferPtr[0] == UNICODE_CR) {
733 Warning (SourceFile->FileName, SourceFile->LineNum, 0, "carriage return found in quoted string", "%S", Start);
734 break;
735 } else if (SourceFile->FileBufferPtr[0] == UNICODE_DOUBLE_QUOTE) {
736 break;
737 }
738
739 SourceFile->FileBufferPtr++;
740 Len++;
741 }
742
743 if (SourceFile->FileBufferPtr[0] != UNICODE_DOUBLE_QUOTE) {
744 Warning (
745 SourceFile->FileName,
746 SourceFile->LineNum,
747 0,
748 "missing closing quote on printable language name string",
749 "%S",
750 Start
751 );
752 } else {
753 SourceFile->FileBufferPtr++;
754 }
755 //
756 // Now allocate memory for the string and save it off
757 //
758 String = (WCHAR *) malloc ((Len + 1) * sizeof (WCHAR));
759 if (String == NULL) {
760 Error (NULL, 0, 0, "memory allocation failed", NULL);
761 return NULL;
762 }
763 //
764 // Copy the string from the file buffer to the local copy.
765 // We do no reformatting of it whatsoever at this point.
766 //
767 Ptr = String;
768 while (Len > 0) {
769 *Ptr = *Start;
770 Start++;
771 Ptr++;
772 Len--;
773 }
774
775 *Ptr = 0;
776 //
777 // Now format the string to convert \wide and \narrow controls
778 //
779 StringDBFormatString (String);
780 return String;
781 }
782
783 static
784 WCHAR *
785 GetQuotedString (
786 SOURCE_FILE *SourceFile,
787 BOOLEAN Optional
788 )
789 {
790 WCHAR *String;
791 WCHAR *Start;
792 WCHAR *Ptr;
793 UINT32 Len;
794 BOOLEAN PreviousBackslash;
795
796 if (SourceFile->FileBufferPtr[0] != UNICODE_DOUBLE_QUOTE) {
797 if (!Optional) {
798 Error (SourceFile->FileName, SourceFile->LineNum, 0, "expected quoted string", "%S", SourceFile->FileBufferPtr);
799 }
800
801 return NULL;
802 }
803
804 Len = 0;
805 SourceFile->FileBufferPtr++;
806 Start = Ptr = SourceFile->FileBufferPtr;
807 PreviousBackslash = FALSE;
808 while (!EndOfFile (SourceFile)) {
809 if ((SourceFile->FileBufferPtr[0] == UNICODE_DOUBLE_QUOTE) && (!PreviousBackslash)) {
810 break;
811 } else if (SourceFile->FileBufferPtr[0] == UNICODE_CR) {
812 Warning (SourceFile->FileName, SourceFile->LineNum, 0, "carriage return found in quoted string", "%S", Start);
813 PreviousBackslash = FALSE;
814 } else if (SourceFile->FileBufferPtr[0] == UNICODE_BACKSLASH) {
815 PreviousBackslash = TRUE;
816 } else {
817 PreviousBackslash = FALSE;
818 }
819
820 SourceFile->FileBufferPtr++;
821 Len++;
822 }
823
824 if (SourceFile->FileBufferPtr[0] != UNICODE_DOUBLE_QUOTE) {
825 Warning (SourceFile->FileName, SourceFile->LineNum, 0, "missing closing quote on string", "%S", Start);
826 } else {
827 SourceFile->FileBufferPtr++;
828 }
829 //
830 // Now allocate memory for the string and save it off
831 //
832 String = (WCHAR *) malloc ((Len + 1) * sizeof (WCHAR));
833 if (String == NULL) {
834 Error (NULL, 0, 0, "memory allocation failed", NULL);
835 return NULL;
836 }
837 //
838 // Copy the string from the file buffer to the local copy.
839 // We do no reformatting of it whatsoever at this point.
840 //
841 Ptr = String;
842 while (Len > 0) {
843 *Ptr = *Start;
844 Start++;
845 Ptr++;
846 Len--;
847 }
848
849 *Ptr = 0;
850 return String;
851 }
852 //
853 // Parse:
854 // #string STR_ID_NAME
855 //
856 // All we can do is call the string database to add the string identifier. Unfortunately
857 // he'll have to keep track of the last identifier we added.
858 //
859 static
860 void
861 ProcessTokenString (
862 SOURCE_FILE *SourceFile
863 )
864 {
865 WCHAR StringIdentifier[MAX_STRING_IDENTIFIER_NAME];
866 UINT16 StringId;
867 //
868 // Extract the string identifier name and add it to the database.
869 //
870 if (GetStringIdentifierName (SourceFile, StringIdentifier, sizeof (StringIdentifier)) > 0) {
871 StringId = STRING_ID_INVALID;
872 StringDBAddStringIdentifier (StringIdentifier, &StringId, 0);
873 } else {
874 //
875 // Error recovery -- skip to the next #
876 //
877 SourceFile->SkipToHash = TRUE;
878 }
879 }
880
881 static
882 BOOLEAN
883 EndOfFile (
884 SOURCE_FILE *SourceFile
885 )
886 {
887 //
888 // The file buffer pointer will typically get updated before the End-of-file flag in the
889 // source file structure, so check it first.
890 //
891 if (SourceFile->FileBufferPtr >= SourceFile->FileBuffer + SourceFile->FileSize / sizeof (WCHAR)) {
892 SourceFile->EndOfFile = TRUE;
893 return TRUE;
894 }
895
896 if (SourceFile->EndOfFile) {
897 return TRUE;
898 }
899
900 return FALSE;
901 }
902
903 static
904 UINT32
905 GetStringIdentifierName (
906 IN SOURCE_FILE *SourceFile,
907 IN OUT WCHAR *StringIdentifierName,
908 IN UINT32 StringIdentifierNameLen
909 )
910 {
911 UINT32 Len;
912 WCHAR *From;
913 WCHAR *Start;
914
915 //
916 // Skip whitespace
917 //
918 SkipWhiteSpace (SourceFile);
919 if (SourceFile->EndOfFile) {
920 Error (SourceFile->FileName, SourceFile->LineNum, 0, "end-of-file encountered", "expected string identifier");
921 return 0;
922 }
923 //
924 // Verify first character of name is [A-Za-z]
925 //
926 Len = 0;
927 StringIdentifierNameLen /= 2;
928 From = SourceFile->FileBufferPtr;
929 Start = SourceFile->FileBufferPtr;
930 if (((SourceFile->FileBufferPtr[0] >= UNICODE_A) && (SourceFile->FileBufferPtr[0] <= UNICODE_Z)) ||
931 ((SourceFile->FileBufferPtr[0] >= UNICODE_z) && (SourceFile->FileBufferPtr[0] <= UNICODE_z))
932 ) {
933 //
934 // Do nothing
935 //
936 } else {
937 Error (SourceFile->FileName, SourceFile->LineNum, 0, "invalid character in string identifier name", "%S", Start);
938 return 0;
939 }
940
941 while (!EndOfFile (SourceFile)) {
942 if (((SourceFile->FileBufferPtr[0] >= UNICODE_A) && (SourceFile->FileBufferPtr[0] <= UNICODE_Z)) ||
943 ((SourceFile->FileBufferPtr[0] >= UNICODE_z) && (SourceFile->FileBufferPtr[0] <= UNICODE_z)) ||
944 ((SourceFile->FileBufferPtr[0] >= UNICODE_0) && (SourceFile->FileBufferPtr[0] <= UNICODE_9)) ||
945 (SourceFile->FileBufferPtr[0] == UNICODE_UNDERSCORE)
946 ) {
947 Len++;
948 if (Len >= StringIdentifierNameLen) {
949 Error (SourceFile->FileName, SourceFile->LineNum, 0, "string identifier name too long", "%S", Start);
950 return 0;
951 }
952
953 *StringIdentifierName = SourceFile->FileBufferPtr[0];
954 StringIdentifierName++;
955 SourceFile->FileBufferPtr++;
956 } else if (SkipWhiteSpace (SourceFile) == 0) {
957 Error (SourceFile->FileName, SourceFile->LineNum, 0, "invalid string identifier name", "%S", Start);
958 return 0;
959 } else {
960 break;
961 }
962 }
963 //
964 // Terminate the copy of the string.
965 //
966 *StringIdentifierName = 0;
967 return Len;
968 }
969
970 static
971 UINT32
972 GetLanguageIdentifierName (
973 IN SOURCE_FILE *SourceFile,
974 IN OUT WCHAR *LanguageIdentifierName,
975 IN UINT32 LanguageIdentifierNameLen,
976 IN BOOLEAN Optional
977 )
978 {
979 UINT32 Len;
980 WCHAR *From;
981 WCHAR *Start;
982 //
983 // Skip whitespace
984 //
985 SkipWhiteSpace (SourceFile);
986 if (SourceFile->EndOfFile) {
987 if (!Optional) {
988 Error (
989 SourceFile->FileName,
990 SourceFile->LineNum,
991 0,
992 "end-of-file encountered",
993 "expected language identifier"
994 );
995 }
996
997 return 0;
998 }
999 //
1000 // This function is called to optionally get a language identifier name in:
1001 // #string STR_ID eng "the string"
1002 // If it's optional, and we find a double-quote, then return now.
1003 //
1004 if (Optional) {
1005 if (*SourceFile->FileBufferPtr == UNICODE_DOUBLE_QUOTE) {
1006 return 0;
1007 }
1008 }
1009
1010 Len = 0;
1011 LanguageIdentifierNameLen /= 2;
1012 //
1013 // Internal error if we weren't given at least 4 WCHAR's to work with.
1014 //
1015 if (LanguageIdentifierNameLen < LANGUAGE_IDENTIFIER_NAME_LEN + 1) {
1016 Error (
1017 SourceFile->FileName,
1018 SourceFile->LineNum,
1019 0,
1020 "app error -- language identifier name length is invalid",
1021 NULL
1022 );
1023 }
1024
1025 From = SourceFile->FileBufferPtr;
1026 Start = SourceFile->FileBufferPtr;
1027 while (!EndOfFile (SourceFile)) {
1028 if (((SourceFile->FileBufferPtr[0] >= UNICODE_a) && (SourceFile->FileBufferPtr[0] <= UNICODE_z))) {
1029 Len++;
1030 if (Len > LANGUAGE_IDENTIFIER_NAME_LEN) {
1031 Error (SourceFile->FileName, SourceFile->LineNum, 0, "language identifier name too long", "%S", Start);
1032 return 0;
1033 }
1034
1035 *LanguageIdentifierName = SourceFile->FileBufferPtr[0];
1036 SourceFile->FileBufferPtr++;
1037 LanguageIdentifierName++;
1038 } else if (!IsWhiteSpace (SourceFile)) {
1039 Error (SourceFile->FileName, SourceFile->LineNum, 0, "invalid language identifier name", "%S", Start);
1040 return 0;
1041 } else {
1042 break;
1043 }
1044 }
1045 //
1046 // Terminate the copy of the string.
1047 //
1048 *LanguageIdentifierName = 0;
1049 return Len;
1050 }
1051
1052 static
1053 void
1054 ProcessTokenInclude (
1055 SOURCE_FILE *SourceFile
1056 )
1057 {
1058 INT8 IncludeFileName[MAX_PATH];
1059 INT8 *To;
1060 UINT32 Len;
1061 BOOLEAN ReportedError;
1062 SOURCE_FILE IncludedSourceFile;
1063
1064 ReportedError = FALSE;
1065 if (SkipWhiteSpace (SourceFile) == 0) {
1066 Warning (SourceFile->FileName, SourceFile->LineNum, 0, "expected whitespace following #include keyword", NULL);
1067 }
1068 //
1069 // Should be quoted file name
1070 //
1071 if (SourceFile->FileBufferPtr[0] != UNICODE_DOUBLE_QUOTE) {
1072 Error (SourceFile->FileName, SourceFile->LineNum, 0, "expected quoted include file name", NULL);
1073 goto FailDone;
1074 }
1075
1076 SourceFile->FileBufferPtr++;
1077 //
1078 // Copy the filename as ascii to our local string
1079 //
1080 To = IncludeFileName;
1081 Len = 0;
1082 while (!EndOfFile (SourceFile)) {
1083 if ((SourceFile->FileBufferPtr[0] == UNICODE_CR) || (SourceFile->FileBufferPtr[0] == UNICODE_LF)) {
1084 Error (SourceFile->FileName, SourceFile->LineNum, 0, "end-of-line found in quoted include file name", NULL);
1085 goto FailDone;
1086 }
1087
1088 if (SourceFile->FileBufferPtr[0] == UNICODE_DOUBLE_QUOTE) {
1089 SourceFile->FileBufferPtr++;
1090 break;
1091 }
1092 //
1093 // If too long, then report the error once and process until the closing quote
1094 //
1095 Len++;
1096 if (!ReportedError && (Len >= sizeof (IncludeFileName))) {
1097 Error (SourceFile->FileName, SourceFile->LineNum, 0, "length of include file name exceeds limit", NULL);
1098 ReportedError = TRUE;
1099 }
1100
1101 if (!ReportedError) {
1102 *To = UNICODE_TO_ASCII (SourceFile->FileBufferPtr[0]);
1103 To++;
1104 }
1105
1106 SourceFile->FileBufferPtr++;
1107 }
1108
1109 if (!ReportedError) {
1110 *To = 0;
1111 memset ((char *) &IncludedSourceFile, 0, sizeof (SOURCE_FILE));
1112 strcpy (IncludedSourceFile.FileName, IncludeFileName);
1113 IncludedSourceFile.ControlCharacter = DEFAULT_CONTROL_CHARACTER;
1114 ProcessIncludeFile (&IncludedSourceFile, SourceFile);
1115 //
1116 // printf ("including file '%s'\n", IncludeFileName);
1117 //
1118 }
1119
1120 return ;
1121 FailDone:
1122 //
1123 // Error recovery -- skip to next #
1124 //
1125 SourceFile->SkipToHash = TRUE;
1126 }
1127
1128 static
1129 void
1130 ProcessTokenScope (
1131 SOURCE_FILE *SourceFile
1132 )
1133 {
1134 WCHAR StringIdentifier[MAX_STRING_IDENTIFIER_NAME];
1135 //
1136 // Extract the scope name
1137 //
1138 if (GetStringIdentifierName (SourceFile, StringIdentifier, sizeof (StringIdentifier)) > 0) {
1139 StringDBSetScope (StringIdentifier);
1140 }
1141 }
1142 //
1143 // Parse: #langdef eng "English"
1144 // #langdef chn "\wideChinese"
1145 //
1146 static
1147 void
1148 ProcessTokenLangDef (
1149 SOURCE_FILE *SourceFile
1150 )
1151 {
1152 WCHAR LanguageIdentifier[MAX_STRING_IDENTIFIER_NAME];
1153 UINT32 Len;
1154 WCHAR *PrintableName;
1155 //
1156 // Extract the 3-character language identifier
1157 //
1158 Len = GetLanguageIdentifierName (SourceFile, LanguageIdentifier, sizeof (LanguageIdentifier), FALSE);
1159 if (Len != LANGUAGE_IDENTIFIER_NAME_LEN) {
1160 Error (SourceFile->FileName, SourceFile->LineNum, 0, "invalid or missing language identifier", NULL);
1161 } else {
1162 //
1163 // Extract the printable name
1164 //
1165 PrintableName = GetPrintableLanguageName (SourceFile);
1166 if (PrintableName != NULL) {
1167 ParserSetPosition (SourceFile->FileName, SourceFile->LineNum);
1168 StringDBAddLanguage (LanguageIdentifier, PrintableName);
1169 free (PrintableName);
1170 return ;
1171 }
1172 }
1173 //
1174 // Error recovery -- skip to next #
1175 //
1176 SourceFile->SkipToHash = TRUE;
1177 }
1178
1179 static
1180 BOOLEAN
1181 ApparentQuotedString (
1182 SOURCE_FILE *SourceFile
1183 )
1184 {
1185 WCHAR *Ptr;
1186 //
1187 // See if the first and last nonblank characters on the line are double quotes
1188 //
1189 for (Ptr = SourceFile->FileBufferPtr; *Ptr && (*Ptr == UNICODE_SPACE); Ptr++)
1190 ;
1191 if (*Ptr != UNICODE_DOUBLE_QUOTE) {
1192 return FALSE;
1193 }
1194
1195 while (*Ptr) {
1196 Ptr++;
1197 }
1198
1199 Ptr--;
1200 for (; *Ptr && (*Ptr == UNICODE_SPACE); Ptr--)
1201 ;
1202 if (*Ptr != UNICODE_DOUBLE_QUOTE) {
1203 return FALSE;
1204 }
1205
1206 return TRUE;
1207 }
1208 //
1209 // Parse:
1210 // #language eng "some string " "more string"
1211 //
1212 static
1213 void
1214 ProcessTokenLanguage (
1215 SOURCE_FILE *SourceFile
1216 )
1217 {
1218 WCHAR *String;
1219 WCHAR *SecondString;
1220 WCHAR *TempString;
1221 WCHAR *From;
1222 WCHAR *To;
1223 WCHAR Language[LANGUAGE_IDENTIFIER_NAME_LEN + 1];
1224 UINT32 Len;
1225 BOOLEAN PreviousNewline;
1226 //
1227 // Get the language identifier
1228 //
1229 Language[0] = 0;
1230 Len = GetLanguageIdentifierName (SourceFile, Language, sizeof (Language), TRUE);
1231 if (Len != LANGUAGE_IDENTIFIER_NAME_LEN) {
1232 Error (SourceFile->FileName, SourceFile->LineNum, 0, "invalid or missing language identifier", "%S", Language);
1233 SourceFile->SkipToHash = TRUE;
1234 return ;
1235 }
1236 //
1237 // Extract the string value. It's either a quoted string that starts on the current line, or
1238 // an unquoted string that starts on the following line and continues until the next control
1239 // character in column 1.
1240 // Look ahead to find a quote or a newline
1241 //
1242 if (SkipTo (SourceFile, UNICODE_DOUBLE_QUOTE, TRUE)) {
1243 String = GetQuotedString (SourceFile, FALSE);
1244 if (String != NULL) {
1245 //
1246 // Set the position in the file of where we are parsing for error
1247 // reporting purposes. Then start looking ahead for additional
1248 // quoted strings, and concatenate them until we get a failure
1249 // back from the string parser.
1250 //
1251 Len = StrLen (String) + 1;
1252 ParserSetPosition (SourceFile->FileName, SourceFile->LineNum);
1253 do {
1254 SkipWhiteSpace (SourceFile);
1255 SecondString = GetQuotedString (SourceFile, TRUE);
1256 if (SecondString != NULL) {
1257 Len += StrLen (SecondString);
1258 TempString = (WCHAR *) malloc (Len * sizeof (WCHAR));
1259 if (TempString == NULL) {
1260 Error (NULL, 0, 0, "application error", "failed to allocate memory");
1261 return ;
1262 }
1263
1264 StrCpy (TempString, String);
1265 StrCat (TempString, SecondString);
1266 free (String);
1267 free (SecondString);
1268 String = TempString;
1269 }
1270 } while (SecondString != NULL);
1271 StringDBAddString (Language, NULL, NULL, String, TRUE, 0);
1272 free (String);
1273 } else {
1274 //
1275 // Error was reported at lower level. Error recovery mode.
1276 //
1277 SourceFile->SkipToHash = TRUE;
1278 }
1279 } else {
1280 if (!mGlobals.UnquotedStrings) {
1281 //
1282 // They're using unquoted strings. If the next non-blank character is a double quote, and the
1283 // last non-blank character on the line is a double quote, then more than likely they're using
1284 // quotes, so they need to put the quoted string on the end of the previous line
1285 //
1286 if (ApparentQuotedString (SourceFile)) {
1287 Warning (
1288 SourceFile->FileName,
1289 SourceFile->LineNum,
1290 0,
1291 "unexpected quoted string on line",
1292 "specify -uqs option if necessary"
1293 );
1294 }
1295 }
1296 //
1297 // Found end-of-line (hopefully). Skip over it and start taking in characters
1298 // until we find a control character at the start of a line.
1299 //
1300 Len = 0;
1301 From = SourceFile->FileBufferPtr;
1302 PreviousNewline = FALSE;
1303 while (!EndOfFile (SourceFile)) {
1304 if (SourceFile->FileBufferPtr[0] == UNICODE_LF) {
1305 PreviousNewline = TRUE;
1306 SourceFile->LineNum++;
1307 } else {
1308 Len++;
1309 if (PreviousNewline && (SourceFile->FileBufferPtr[0] == SourceFile->ControlCharacter)) {
1310 break;
1311 }
1312
1313 PreviousNewline = FALSE;
1314 }
1315
1316 SourceFile->FileBufferPtr++;
1317 }
1318
1319 if ((Len == 0) && EndOfFile (SourceFile)) {
1320 Error (SourceFile->FileName, SourceFile->LineNum, 0, "unexpected end of file", NULL);
1321 SourceFile->SkipToHash = TRUE;
1322 return ;
1323 }
1324 //
1325 // Now allocate a buffer, copy the characters, and add the string.
1326 //
1327 String = (WCHAR *) malloc ((Len + 1) * sizeof (WCHAR));
1328 if (String == NULL) {
1329 Error (NULL, 0, 0, "application error", "failed to allocate memory");
1330 return ;
1331 }
1332
1333 To = String;
1334 while (From < SourceFile->FileBufferPtr) {
1335 switch (*From) {
1336 case UNICODE_LF:
1337 case 0:
1338 break;
1339
1340 default:
1341 *To = *From;
1342 To++;
1343 break;
1344 }
1345
1346 From++;
1347 }
1348
1349 //
1350 // String[Len] = 0;
1351 //
1352 *To = 0;
1353 StringDBAddString (Language, NULL, NULL, String, TRUE, 0);
1354 }
1355 }
1356
1357 static
1358 BOOLEAN
1359 IsWhiteSpace (
1360 SOURCE_FILE *SourceFile
1361 )
1362 {
1363 switch (SourceFile->FileBufferPtr[0]) {
1364 case UNICODE_NULL:
1365 case UNICODE_CR:
1366 case UNICODE_SPACE:
1367 case UNICODE_TAB:
1368 case UNICODE_LF:
1369 return TRUE;
1370
1371 default:
1372 return FALSE;
1373 }
1374 }
1375
1376 static
1377 UINT32
1378 SkipWhiteSpace (
1379 SOURCE_FILE *SourceFile
1380 )
1381 {
1382 UINT32 Count;
1383
1384 Count = 0;
1385 while (!EndOfFile (SourceFile)) {
1386 Count++;
1387 switch (*SourceFile->FileBufferPtr) {
1388 case UNICODE_NULL:
1389 case UNICODE_CR:
1390 case UNICODE_SPACE:
1391 case UNICODE_TAB:
1392 SourceFile->FileBufferPtr++;
1393 break;
1394
1395 case UNICODE_LF:
1396 SourceFile->FileBufferPtr++;
1397 SourceFile->LineNum++;
1398 if (mGlobals.Verbose) {
1399 printf ("%d: %S\n", SourceFile->LineNum, SourceFile->FileBufferPtr);
1400 }
1401 break;
1402
1403 default:
1404 return Count - 1;
1405 }
1406 }
1407 //
1408 // Some tokens require trailing whitespace. If we're at the end of the
1409 // file, then we count that as well.
1410 //
1411 if ((Count == 0) && (EndOfFile (SourceFile))) {
1412 Count++;
1413 }
1414
1415 return Count;
1416 }
1417
1418 static
1419 UINT32
1420 wstrcmp (
1421 WCHAR *Buffer,
1422 WCHAR *Str
1423 )
1424 {
1425 UINT32 Len;
1426
1427 Len = 0;
1428 while (*Str == *Buffer) {
1429 Buffer++;
1430 Str++;
1431 Len++;
1432 }
1433
1434 if (*Str) {
1435 return 0;
1436 }
1437
1438 return Len;
1439 }
1440 //
1441 // Given a filename, try to find it along the include paths.
1442 //
1443 static
1444 FILE *
1445 FindFile (
1446 IN INT8 *FileName,
1447 OUT INT8 *FoundFileName,
1448 IN UINT32 FoundFileNameLen
1449 )
1450 {
1451 FILE *Fptr;
1452 TEXT_STRING_LIST *List;
1453
1454 //
1455 // Traverse the list of paths and try to find the file
1456 //
1457 List = mGlobals.IncludePaths;
1458 while (List != NULL) {
1459 //
1460 // Put the path and filename together
1461 //
1462 if (strlen (List->Str) + strlen (FileName) + 1 > FoundFileNameLen) {
1463 Error (PROGRAM_NAME, 0, 0, NULL, "internal error - cannot concatenate path+filename");
1464 return NULL;
1465 }
1466 //
1467 // Append the filename to this include path and try to open the file.
1468 //
1469 strcpy (FoundFileName, List->Str);
1470 strcat (FoundFileName, FileName);
1471 if ((Fptr = fopen (FoundFileName, "rb")) != NULL) {
1472 //
1473 // Return the file pointer
1474 //
1475 return Fptr;
1476 }
1477
1478 List = List->Next;
1479 }
1480 //
1481 // Not found
1482 //
1483 FoundFileName[0] = 0;
1484 return NULL;
1485 }
1486 //
1487 // Process the command-line arguments
1488 //
1489 static
1490 STATUS
1491 ProcessArgs (
1492 int Argc,
1493 char *Argv[]
1494 )
1495 {
1496 TEXT_STRING_LIST *NewList;
1497 //
1498 // Clear our globals
1499 //
1500 memset ((char *) &mGlobals, 0, sizeof (mGlobals));
1501 strcpy (mGlobals.BaseName, DEFAULT_BASE_NAME);
1502 //
1503 // Skip program name
1504 //
1505 Argc--;
1506 Argv++;
1507
1508 if (Argc == 0) {
1509 Usage ();
1510 return STATUS_ERROR;
1511 }
1512
1513 mGlobals.Mode = MODE_UNKNOWN;
1514 //
1515 // Process until no more -args.
1516 //
1517 while ((Argc > 0) && (Argv[0][0] == '-')) {
1518 //
1519 // -parse option
1520 //
1521 if (stricmp (Argv[0], "-parse") == 0) {
1522 if (mGlobals.Mode != MODE_UNKNOWN) {
1523 Error (NULL, 0, 0, "only one of -parse/-scan/-dump allowed", NULL);
1524 return STATUS_ERROR;
1525 }
1526
1527 mGlobals.Mode = MODE_PARSE;
1528 //
1529 // -scan option
1530 //
1531 } else if (stricmp (Argv[0], "-scan") == 0) {
1532 if (mGlobals.Mode != MODE_UNKNOWN) {
1533 Error (NULL, 0, 0, "only one of -parse/-scan/-dump allowed", NULL);
1534 return STATUS_ERROR;
1535 }
1536
1537 mGlobals.Mode = MODE_SCAN;
1538 //
1539 // -vscan verbose scanning option
1540 //
1541 } else if (stricmp (Argv[0], "-vscan") == 0) {
1542 mGlobals.VerboseScan = TRUE;
1543 //
1544 // -dump option
1545 //
1546 } else if (stricmp (Argv[0], "-dump") == 0) {
1547 if (mGlobals.Mode != MODE_UNKNOWN) {
1548 Error (NULL, 0, 0, "only one of -parse/-scan/-dump allowed", NULL);
1549 return STATUS_ERROR;
1550 }
1551
1552 mGlobals.Mode = MODE_DUMP;
1553 } else if (stricmp (Argv[0], "-uqs") == 0) {
1554 mGlobals.UnquotedStrings = TRUE;
1555 //
1556 // -i path add include search path when parsing
1557 //
1558 } else if (stricmp (Argv[0], "-i") == 0) {
1559 //
1560 // check for one more arg
1561 //
1562 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1563 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing include path");
1564 return STATUS_ERROR;
1565 }
1566 //
1567 // Allocate memory for a new list element, fill it in, and
1568 // add it to our list of include paths. Always make sure it
1569 // has a "\" on the end of it.
1570 //
1571 NewList = malloc (sizeof (TEXT_STRING_LIST));
1572 if (NewList == NULL) {
1573 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1574 return STATUS_ERROR;
1575 }
1576
1577 memset ((char *) NewList, 0, sizeof (TEXT_STRING_LIST));
1578 NewList->Str = malloc (strlen (Argv[1]) + 2);
1579 if (NewList->Str == NULL) {
1580 free (NewList);
1581 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1582 return STATUS_ERROR;
1583 }
1584
1585 strcpy (NewList->Str, Argv[1]);
1586 if (NewList->Str[strlen (NewList->Str) - 1] != '\\') {
1587 strcat (NewList->Str, "\\");
1588 }
1589 //
1590 // Add it to our linked list
1591 //
1592 if (mGlobals.IncludePaths == NULL) {
1593 mGlobals.IncludePaths = NewList;
1594 } else {
1595 mGlobals.LastIncludePath->Next = NewList;
1596 }
1597
1598 mGlobals.LastIncludePath = NewList;
1599 Argc--;
1600 Argv++;
1601 } else if (stricmp (Argv[0], "-if") == 0) {
1602 //
1603 // Indirection file -- check for one more arg
1604 //
1605 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1606 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing indirection file name");
1607 return STATUS_ERROR;
1608 }
1609 //
1610 // Allocate memory for a new list element, fill it in, and
1611 // add it to our list of include paths. Always make sure it
1612 // has a "\" on the end of it.
1613 //
1614 NewList = malloc (sizeof (TEXT_STRING_LIST));
1615 if (NewList == NULL) {
1616 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1617 return STATUS_ERROR;
1618 }
1619
1620 memset ((char *) NewList, 0, sizeof (TEXT_STRING_LIST));
1621 NewList->Str = malloc (strlen (Argv[1]) + 1);
1622 if (NewList->Str == NULL) {
1623 free (NewList);
1624 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1625 return STATUS_ERROR;
1626 }
1627
1628 strcpy (NewList->Str, Argv[1]);
1629 //
1630 // Add it to our linked list
1631 //
1632 if (mGlobals.IndirectionFileName == NULL) {
1633 mGlobals.IndirectionFileName = NewList;
1634 } else {
1635 mGlobals.LastIndirectionFileName->Next = NewList;
1636 }
1637
1638 mGlobals.LastIndirectionFileName = NewList;
1639 Argc--;
1640 Argv++;
1641 } else if (stricmp (Argv[0], "-db") == 0) {
1642 //
1643 // -db option to specify a database file.
1644 // Check for one more arg (the database file name)
1645 //
1646 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1647 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing database file name");
1648 return STATUS_ERROR;
1649 }
1650
1651 NewList = malloc (sizeof (TEXT_STRING_LIST));
1652 if (NewList == NULL) {
1653 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1654 return STATUS_ERROR;
1655 }
1656
1657 memset ((char *) NewList, 0, sizeof (TEXT_STRING_LIST));
1658 NewList->Str = malloc (strlen (Argv[1]) + 1);
1659 if (NewList->Str == NULL) {
1660 free (NewList);
1661 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1662 return STATUS_ERROR;
1663 }
1664
1665 strcpy (NewList->Str, Argv[1]);
1666 //
1667 // Add it to our linked list
1668 //
1669 if (mGlobals.DatabaseFileName == NULL) {
1670 mGlobals.DatabaseFileName = NewList;
1671 } else {
1672 mGlobals.LastDatabaseFileName->Next = NewList;
1673 }
1674
1675 mGlobals.LastDatabaseFileName = NewList;
1676 Argc--;
1677 Argv++;
1678 } else if (stricmp (Argv[0], "-ou") == 0) {
1679 //
1680 // -ou option to specify an output unicode file to
1681 // which we can dump our database.
1682 //
1683 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1684 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing database dump output file name");
1685 return STATUS_ERROR;
1686 }
1687
1688 if (mGlobals.DumpUFileName[0] == 0) {
1689 strcpy (mGlobals.DumpUFileName, Argv[1]);
1690 } else {
1691 Error (PROGRAM_NAME, 0, 0, Argv[1], "-ou option already specified with '%s'", mGlobals.DumpUFileName);
1692 return STATUS_ERROR;
1693 }
1694
1695 Argc--;
1696 Argv++;
1697 } else if (stricmp (Argv[0], "-hpk") == 0) {
1698 //
1699 // -hpk option to create an HII export pack of the input database file
1700 //
1701 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1702 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing raw string data dump output file name");
1703 return STATUS_ERROR;
1704 }
1705
1706 if (mGlobals.HiiExportPackFileName[0] == 0) {
1707 strcpy (mGlobals.HiiExportPackFileName, Argv[1]);
1708 } else {
1709 Error (PROGRAM_NAME, 0, 0, Argv[1], "-or option already specified with '%s'", mGlobals.HiiExportPackFileName);
1710 return STATUS_ERROR;
1711 }
1712
1713 Argc--;
1714 Argv++;
1715 } else if ((stricmp (Argv[0], "-?") == 0) || (stricmp (Argv[0], "-h") == 0)) {
1716 Usage ();
1717 return STATUS_ERROR;
1718 } else if (stricmp (Argv[0], "-v") == 0) {
1719 mGlobals.Verbose = 1;
1720 } else if (stricmp (Argv[0], "-vdbw") == 0) {
1721 mGlobals.VerboseDatabaseWrite = 1;
1722 } else if (stricmp (Argv[0], "-vdbr") == 0) {
1723 mGlobals.VerboseDatabaseRead = 1;
1724 } else if (stricmp (Argv[0], "-newdb") == 0) {
1725 mGlobals.NewDatabase = 1;
1726 } else if (stricmp (Argv[0], "-ignorenotfound") == 0) {
1727 mGlobals.IgnoreNotFound = 1;
1728 } else if (stricmp (Argv[0], "-oc") == 0) {
1729 //
1730 // check for one more arg
1731 //
1732 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1733 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing output C filename");
1734 return STATUS_ERROR;
1735 }
1736
1737 strcpy (mGlobals.StringCFileName, Argv[1]);
1738 Argc--;
1739 Argv++;
1740 } else if (stricmp (Argv[0], "-bn") == 0) {
1741 //
1742 // check for one more arg
1743 //
1744 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1745 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing base name");
1746 Usage ();
1747 return STATUS_ERROR;
1748 }
1749
1750 strcpy (mGlobals.BaseName, Argv[1]);
1751 Argc--;
1752 Argv++;
1753 } else if (stricmp (Argv[0], "-oh") == 0) {
1754 //
1755 // -oh to specify output .h defines file name
1756 //
1757 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1758 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing output .h filename");
1759 return STATUS_ERROR;
1760 }
1761
1762 strcpy (mGlobals.StringHFileName, Argv[1]);
1763 Argc--;
1764 Argv++;
1765 } else if (stricmp (Argv[0], "-skipext") == 0) {
1766 //
1767 // -skipext to skip scanning of files with certain filename extensions
1768 //
1769 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1770 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing filename extension");
1771 return STATUS_ERROR;
1772 }
1773 //
1774 // Allocate memory for a new list element, fill it in, and
1775 // add it to our list of excluded extensions. Always make sure it
1776 // has a "." as the first character.
1777 //
1778 NewList = malloc (sizeof (TEXT_STRING_LIST));
1779 if (NewList == NULL) {
1780 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1781 return STATUS_ERROR;
1782 }
1783
1784 memset ((char *) NewList, 0, sizeof (TEXT_STRING_LIST));
1785 NewList->Str = malloc (strlen (Argv[1]) + 2);
1786 if (NewList->Str == NULL) {
1787 free (NewList);
1788 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1789 return STATUS_ERROR;
1790 }
1791
1792 if (Argv[1][0] == '.') {
1793 strcpy (NewList->Str, Argv[1]);
1794 } else {
1795 NewList->Str[0] = '.';
1796 strcpy (NewList->Str + 1, Argv[1]);
1797 }
1798 //
1799 // Add it to our linked list
1800 //
1801 if (mGlobals.SkipExt == NULL) {
1802 mGlobals.SkipExt = NewList;
1803 } else {
1804 mGlobals.LastSkipExt->Next = NewList;
1805 }
1806
1807 mGlobals.LastSkipExt = NewList;
1808 Argc--;
1809 Argv++;
1810 } else if (stricmp (Argv[0], "-lang") == 0) {
1811 //
1812 // "-lang eng" or "-lang spa+cat" to only output certain languages
1813 //
1814 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1815 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing language name");
1816 Usage ();
1817 return STATUS_ERROR;
1818 }
1819
1820 if (AddCommandLineLanguage (Argv[1]) != STATUS_SUCCESS) {
1821 return STATUS_ERROR;
1822 }
1823
1824 Argc--;
1825 Argv++;
1826 } else if (stricmp (Argv[0], "-od") == 0) {
1827 //
1828 // Output database file name -- check for another arg
1829 //
1830 if ((Argc <= 1) || (Argv[1][0] == '-')) {
1831 Error (PROGRAM_NAME, 0, 0, Argv[0], "missing output database file name");
1832 return STATUS_ERROR;
1833 }
1834
1835 strcpy (mGlobals.OutputDatabaseFileName, Argv[1]);
1836 Argv++;
1837 Argc--;
1838 } else {
1839 //
1840 // Unrecognized arg
1841 //
1842 Error (PROGRAM_NAME, 0, 0, Argv[0], "unrecognized option");
1843 Usage ();
1844 return STATUS_ERROR;
1845 }
1846
1847 Argv++;
1848 Argc--;
1849 }
1850 //
1851 // Make sure they specified the mode parse/scan/dump
1852 //
1853 if (mGlobals.Mode == MODE_UNKNOWN) {
1854 Error (NULL, 0, 0, "must specify one of -parse/-scan/-dump", NULL);
1855 return STATUS_ERROR;
1856 }
1857 //
1858 // All modes require a database filename
1859 //
1860 if (mGlobals.DatabaseFileName == 0) {
1861 Error (NULL, 0, 0, "must specify a database filename using -db DbFileName", NULL);
1862 Usage ();
1863 return STATUS_ERROR;
1864 }
1865 //
1866 // If dumping the database file, then return immediately if all
1867 // parameters check out.
1868 //
1869 if (mGlobals.Mode == MODE_DUMP) {
1870 //
1871 // Not much use if they didn't specify -oh or -oc or -ou or -hpk
1872 //
1873 if ((mGlobals.DumpUFileName[0] == 0) &&
1874 (mGlobals.StringHFileName[0] == 0) &&
1875 (mGlobals.StringCFileName[0] == 0) &&
1876 (mGlobals.HiiExportPackFileName[0] == 0)
1877 ) {
1878 Error (NULL, 0, 0, "-dump without -oc/-oh/-ou/-hpk is a NOP", NULL);
1879 return STATUS_ERROR;
1880 }
1881
1882 return STATUS_SUCCESS;
1883 }
1884 //
1885 // Had to specify source string file and output string defines header filename.
1886 //
1887 if (mGlobals.Mode == MODE_SCAN) {
1888 if (Argc < 1) {
1889 Error (PROGRAM_NAME, 0, 0, NULL, "must specify at least one source file to scan with -scan");
1890 Usage ();
1891 return STATUS_ERROR;
1892 }
1893 //
1894 // Get the list of filenames
1895 //
1896 while (Argc > 0) {
1897 NewList = malloc (sizeof (TEXT_STRING_LIST));
1898 if (NewList == NULL) {
1899 Error (PROGRAM_NAME, 0, 0, "memory allocation failure", NULL);
1900 return STATUS_ERROR;
1901 }
1902
1903 memset (NewList, 0, sizeof (TEXT_STRING_LIST));
1904 NewList->Str = (UINT8 *) malloc (strlen (Argv[0]) + 1);
1905 if (NewList->Str == NULL) {
1906 Error (PROGRAM_NAME, 0, 0, "memory allocation failure", NULL);
1907 return STATUS_ERROR;
1908 }
1909
1910 strcpy (NewList->Str, Argv[0]);
1911 if (mGlobals.ScanFileName == NULL) {
1912 mGlobals.ScanFileName = NewList;
1913 } else {
1914 mGlobals.LastScanFileName->Next = NewList;
1915 }
1916
1917 mGlobals.LastScanFileName = NewList;
1918 Argc--;
1919 Argv++;
1920 }
1921 } else {
1922 //
1923 // Parse mode -- must specify an input unicode file name
1924 //
1925 if (Argc < 1) {
1926 Error (PROGRAM_NAME, 0, 0, NULL, "must specify input unicode string file name with -parse");
1927 Usage ();
1928 return STATUS_ERROR;
1929 }
1930
1931 strcpy (mGlobals.SourceFiles.FileName, Argv[0]);
1932 }
1933
1934 return STATUS_SUCCESS;
1935 }
1936 //
1937 // Found "-lang eng,spa+cat" on the command line. Parse the
1938 // language list and save the setting for later processing.
1939 //
1940 static
1941 STATUS
1942 AddCommandLineLanguage (
1943 IN INT8 *Language
1944 )
1945 {
1946 WCHAR_STRING_LIST *WNewList;
1947 WCHAR *From;
1948 WCHAR *To;
1949 //
1950 // Keep processing the input string until we find the end.
1951 //
1952 while (*Language) {
1953 //
1954 // Allocate memory for a new list element, fill it in, and
1955 // add it to our list.
1956 //
1957 WNewList = MALLOC (sizeof (WCHAR_STRING_LIST));
1958 if (WNewList == NULL) {
1959 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1960 return STATUS_ERROR;
1961 }
1962
1963 memset ((char *) WNewList, 0, sizeof (WCHAR_STRING_LIST));
1964 WNewList->Str = malloc ((strlen (Language) + 1) * sizeof (WCHAR));
1965 if (WNewList->Str == NULL) {
1966 free (WNewList);
1967 Error (PROGRAM_NAME, 0, 0, NULL, "memory allocation failure");
1968 return STATUS_ERROR;
1969 }
1970 //
1971 // Copy it as unicode to our new structure. Then remove the
1972 // plus signs in it, and verify each language name is 3 characters
1973 // long. If we find a comma, then we're done with this group, so
1974 // break out.
1975 //
1976 UnicodeSPrint (WNewList->Str, (strlen (Language) + 1) * sizeof (WCHAR), L"%a", Language);
1977 From = To = WNewList->Str;
1978 while (*From) {
1979 if (*From == L',') {
1980 break;
1981 }
1982
1983 if ((StrLen (From) < LANGUAGE_IDENTIFIER_NAME_LEN) ||
1984 (
1985 (From[LANGUAGE_IDENTIFIER_NAME_LEN] != 0) &&
1986 (From[LANGUAGE_IDENTIFIER_NAME_LEN] != UNICODE_PLUS_SIGN) &&
1987 (From[LANGUAGE_IDENTIFIER_NAME_LEN] != L',')
1988 )
1989 ) {
1990 Error (PROGRAM_NAME, 0, 0, Language, "invalid format for language name on command line");
1991 FREE (WNewList->Str);
1992 FREE (WNewList);
1993 return STATUS_ERROR;
1994 }
1995
1996 StrnCpy (To, From, LANGUAGE_IDENTIFIER_NAME_LEN);
1997 To += LANGUAGE_IDENTIFIER_NAME_LEN;
1998 From += LANGUAGE_IDENTIFIER_NAME_LEN;
1999 if (*From == L'+') {
2000 From++;
2001 }
2002 }
2003
2004 *To = 0;
2005 //
2006 // Add it to our linked list
2007 //
2008 if (mGlobals.Language == NULL) {
2009 mGlobals.Language = WNewList;
2010 } else {
2011 mGlobals.LastLanguage->Next = WNewList;
2012 }
2013
2014 mGlobals.LastLanguage = WNewList;
2015 //
2016 // Skip to next entry (comma-separated list)
2017 //
2018 while (*Language) {
2019 if (*Language == L',') {
2020 Language++;
2021 break;
2022 }
2023
2024 Language++;
2025 }
2026 }
2027
2028 return STATUS_SUCCESS;
2029 }
2030 //
2031 // The contents of the text file are expected to be (one per line)
2032 // STRING_IDENTIFIER_NAME ScopeName
2033 // For example:
2034 // STR_ID_MY_FAVORITE_STRING IBM
2035 //
2036 static
2037 STATUS
2038 ParseIndirectionFiles (
2039 TEXT_STRING_LIST *Files
2040 )
2041 {
2042 FILE *Fptr;
2043 INT8 Line[200];
2044 INT8 *StringName;
2045 INT8 *ScopeName;
2046 INT8 *End;
2047 UINT32 LineCount;
2048 WCHAR_MATCHING_STRING_LIST *NewList;
2049
2050 Line[sizeof (Line) - 1] = 0;
2051 Fptr = NULL;
2052 while (Files != NULL) {
2053 Fptr = fopen (Files->Str, "r");
2054 LineCount = 0;
2055 if (Fptr == NULL) {
2056 Error (NULL, 0, 0, Files->Str, "failed to open input indirection file for reading");
2057 return STATUS_ERROR;
2058 }
2059
2060 while (fgets (Line, sizeof (Line), Fptr) != NULL) {
2061 //
2062 // remove terminating newline for error printing purposes.
2063 //
2064 if (Line[strlen (Line) - 1] == '\n') {
2065 Line[strlen (Line) - 1] = 0;
2066 }
2067
2068 LineCount++;
2069 if (Line[sizeof (Line) - 1] != 0) {
2070 Error (Files->Str, LineCount, 0, "line length exceeds maximum supported", NULL);
2071 goto Done;
2072 }
2073
2074 StringName = Line;
2075 while (*StringName && (isspace (*StringName))) {
2076 StringName++;
2077 }
2078
2079 if (*StringName) {
2080 if ((*StringName == '_') || isalpha (*StringName)) {
2081 End = StringName;
2082 while ((*End) && (*End == '_') || (isalnum (*End))) {
2083 End++;
2084 }
2085
2086 if (isspace (*End)) {
2087 *End = 0;
2088 End++;
2089 while (isspace (*End)) {
2090 End++;
2091 }
2092
2093 if (*End) {
2094 ScopeName = End;
2095 while (*End && !isspace (*End)) {
2096 End++;
2097 }
2098
2099 *End = 0;
2100 //
2101 // Add the string name/scope pair
2102 //
2103 NewList = malloc (sizeof (WCHAR_MATCHING_STRING_LIST));
2104 if (NewList == NULL) {
2105 Error (NULL, 0, 0, "memory allocation error", NULL);
2106 goto Done;
2107 }
2108
2109 memset (NewList, 0, sizeof (WCHAR_MATCHING_STRING_LIST));
2110 NewList->Str1 = (WCHAR *) malloc ((strlen (StringName) + 1) * sizeof (WCHAR));
2111 NewList->Str2 = (WCHAR *) malloc ((strlen (ScopeName) + 1) * sizeof (WCHAR));
2112 if ((NewList->Str1 == NULL) || (NewList->Str2 == NULL)) {
2113 Error (NULL, 0, 0, "memory allocation error", NULL);
2114 goto Done;
2115 }
2116
2117 UnicodeSPrint (NewList->Str1, strlen (StringName) + 1, L"%a", StringName);
2118 UnicodeSPrint (NewList->Str2, strlen (ScopeName) + 1, L"%a", ScopeName);
2119 if (mGlobals.IndirectionList == NULL) {
2120 mGlobals.IndirectionList = NewList;
2121 } else {
2122 mGlobals.LastIndirectionList->Next = NewList;
2123 }
2124
2125 mGlobals.LastIndirectionList = NewList;
2126 } else {
2127 Error (Files->Str, LineCount, 0, StringName, "invalid line : expected 'StringIdentifier Scope'");
2128 goto Done;
2129 }
2130 } else {
2131 Error (Files->Str, LineCount, 0, StringName, "invalid line : expected 'StringIdentifier Scope'");
2132 goto Done;
2133 }
2134 } else {
2135 Error (Files->Str, LineCount, 0, StringName, "invalid string identifier");
2136 goto Done;
2137 }
2138 }
2139 }
2140
2141 fclose (Fptr);
2142 Fptr = NULL;
2143 Files = Files->Next;
2144 }
2145
2146 Done:
2147 if (Fptr != NULL) {
2148 fclose (Fptr);
2149 return STATUS_ERROR;
2150 }
2151
2152 return STATUS_SUCCESS;
2153 }
2154
2155 static
2156 STATUS
2157 ScanFiles (
2158 TEXT_STRING_LIST *ScanFiles
2159 )
2160 {
2161 char Line[MAX_LINE_LEN];
2162 FILE *Fptr;
2163 UINT32 LineNum;
2164 char *Cptr;
2165 char *SavePtr;
2166 char *TermPtr;
2167 char *StringTokenPos;
2168 TEXT_STRING_LIST *SList;
2169 BOOLEAN SkipIt;
2170
2171 //
2172 // Put a null-terminator at the end of the line. If we read in
2173 // a line longer than we support, then we can catch it.
2174 //
2175 Line[MAX_LINE_LEN - 1] = 0;
2176 //
2177 // Process each file. If they gave us a skip extension list, then
2178 // skip it if the extension matches.
2179 //
2180 while (ScanFiles != NULL) {
2181 SkipIt = FALSE;
2182 for (SList = mGlobals.SkipExt; SList != NULL; SList = SList->Next) {
2183 if ((strlen (ScanFiles->Str) > strlen (SList->Str)) &&
2184 (strcmp (ScanFiles->Str + strlen (ScanFiles->Str) - strlen (SList->Str), SList->Str) == 0)
2185 ) {
2186 SkipIt = TRUE;
2187 //
2188 // printf ("Match: %s : %s\n", ScanFiles->Str, SList->Str);
2189 //
2190 break;
2191 }
2192 }
2193
2194 if (!SkipIt) {
2195 if (mGlobals.VerboseScan) {
2196 printf ("Scanning %s\n", ScanFiles->Str);
2197 }
2198
2199 Fptr = fopen (ScanFiles->Str, "r");
2200 if (Fptr == NULL) {
2201 Error (NULL, 0, 0, ScanFiles->Str, "failed to open input file for scanning");
2202 return STATUS_ERROR;
2203 }
2204
2205 LineNum = 0;
2206 while (fgets (Line, sizeof (Line), Fptr) != NULL) {
2207 LineNum++;
2208 if (Line[MAX_LINE_LEN - 1] != 0) {
2209 Error (ScanFiles->Str, LineNum, 0, "line length exceeds maximum supported by tool", NULL);
2210 fclose (Fptr);
2211 return STATUS_ERROR;
2212 }
2213 //
2214 // Remove the newline from the input line so we can print a warning message
2215 //
2216 if (Line[strlen (Line) - 1] == '\n') {
2217 Line[strlen (Line) - 1] = 0;
2218 }
2219 //
2220 // Terminate the line at // comments
2221 //
2222 Cptr = strstr (Line, "//");
2223 if (Cptr != NULL) {
2224 *Cptr = 0;
2225 }
2226
2227 Cptr = Line;
2228 while ((Cptr = strstr (Cptr, STRING_TOKEN)) != NULL) {
2229 //
2230 // Found "STRING_TOKEN". Make sure we don't have NUM_STRING_TOKENS or
2231 // something like that. Then make sure it's followed by
2232 // an open parenthesis, a string identifier, and then a closing
2233 // parenthesis.
2234 //
2235 if (mGlobals.VerboseScan) {
2236 printf (" %d: %s", LineNum, Cptr);
2237 }
2238
2239 if (((Cptr == Line) || (!IsValidIdentifierChar (*(Cptr - 1), FALSE))) &&
2240 (!IsValidIdentifierChar (*(Cptr + sizeof (STRING_TOKEN) - 1), FALSE))
2241 ) {
2242 StringTokenPos = Cptr;
2243 SavePtr = Cptr;
2244 Cptr += strlen (STRING_TOKEN);
2245 while (*Cptr && isspace (*Cptr) && (*Cptr != '(')) {
2246 Cptr++;
2247 }
2248
2249 if (*Cptr != '(') {
2250 Warning (ScanFiles->Str, LineNum, 0, StringTokenPos, "expected "STRING_TOKEN "(identifier)");
2251 } else {
2252 //
2253 // Skip over the open-parenthesis and find the next non-blank character
2254 //
2255 Cptr++;
2256 while (isspace (*Cptr)) {
2257 Cptr++;
2258 }
2259
2260 SavePtr = Cptr;
2261 if ((*Cptr == '_') || isalpha (*Cptr)) {
2262 while ((*Cptr == '_') || (isalnum (*Cptr))) {
2263 Cptr++;
2264 }
2265
2266 TermPtr = Cptr;
2267 while (*Cptr && isspace (*Cptr)) {
2268 Cptr++;
2269 }
2270
2271 if (*Cptr != ')') {
2272 Warning (ScanFiles->Str, LineNum, 0, StringTokenPos, "expected "STRING_TOKEN "(identifier)");
2273 }
2274
2275 if (*TermPtr) {
2276 *TermPtr = 0;
2277 Cptr = TermPtr + 1;
2278 } else {
2279 Cptr = TermPtr;
2280 }
2281 //
2282 // Add the string identifier to the list of used strings
2283 //
2284 ParserSetPosition (ScanFiles->Str, LineNum);
2285 StringDBSetStringReferenced (SavePtr, mGlobals.IgnoreNotFound);
2286 if (mGlobals.VerboseScan) {
2287 printf ("...referenced %s", SavePtr);
2288 }
2289 } else {
2290 Warning (ScanFiles->Str, LineNum, 0, StringTokenPos, "expected valid string identifier name");
2291 }
2292 }
2293 } else {
2294 //
2295 // Found it, but it's a substring of something else. Advance our pointer.
2296 //
2297 Cptr++;
2298 }
2299
2300 if (mGlobals.VerboseScan) {
2301 printf ("\n");
2302 }
2303 }
2304 }
2305
2306 fclose (Fptr);
2307 } else {
2308 //
2309 // Skipping this file type
2310 //
2311 if (mGlobals.VerboseScan) {
2312 printf ("Skip scanning of %s\n", ScanFiles->Str);
2313 }
2314 }
2315
2316 ScanFiles = ScanFiles->Next;
2317 }
2318
2319 return STATUS_SUCCESS;
2320 }
2321 //
2322 // Free the global string lists we allocated memory for
2323 //
2324 static
2325 void
2326 FreeLists (
2327 VOID
2328 )
2329 {
2330 TEXT_STRING_LIST *Temp;
2331 WCHAR_STRING_LIST *WTemp;
2332
2333 //
2334 // Traverse the include paths, freeing each
2335 //
2336 while (mGlobals.IncludePaths != NULL) {
2337 Temp = mGlobals.IncludePaths->Next;
2338 free (mGlobals.IncludePaths->Str);
2339 free (mGlobals.IncludePaths);
2340 mGlobals.IncludePaths = Temp;
2341 }
2342 //
2343 // If we did a scan, then free up our
2344 // list of files to scan.
2345 //
2346 while (mGlobals.ScanFileName != NULL) {
2347 Temp = mGlobals.ScanFileName->Next;
2348 free (mGlobals.ScanFileName->Str);
2349 free (mGlobals.ScanFileName);
2350 mGlobals.ScanFileName = Temp;
2351 }
2352 //
2353 // If they gave us a list of filename extensions to
2354 // skip on scan, then free them up.
2355 //
2356 while (mGlobals.SkipExt != NULL) {
2357 Temp = mGlobals.SkipExt->Next;
2358 free (mGlobals.SkipExt->Str);
2359 free (mGlobals.SkipExt);
2360 mGlobals.SkipExt = Temp;
2361 }
2362 //
2363 // Free up any languages specified
2364 //
2365 while (mGlobals.Language != NULL) {
2366 WTemp = mGlobals.Language->Next;
2367 free (mGlobals.Language->Str);
2368 free (mGlobals.Language);
2369 mGlobals.Language = WTemp;
2370 }
2371 //
2372 // Free up our indirection list
2373 //
2374 while (mGlobals.IndirectionList != NULL) {
2375 mGlobals.LastIndirectionList = mGlobals.IndirectionList->Next;
2376 free (mGlobals.IndirectionList->Str1);
2377 free (mGlobals.IndirectionList->Str2);
2378 free (mGlobals.IndirectionList);
2379 mGlobals.IndirectionList = mGlobals.LastIndirectionList;
2380 }
2381
2382 while (mGlobals.IndirectionFileName != NULL) {
2383 mGlobals.LastIndirectionFileName = mGlobals.IndirectionFileName->Next;
2384 free (mGlobals.IndirectionFileName->Str);
2385 free (mGlobals.IndirectionFileName);
2386 mGlobals.IndirectionFileName = mGlobals.LastIndirectionFileName;
2387 }
2388 }
2389
2390 static
2391 BOOLEAN
2392 IsValidIdentifierChar (
2393 INT8 Char,
2394 BOOLEAN FirstChar
2395 )
2396 {
2397 //
2398 // If it's the first character of an identifier, then
2399 // it must be one of [A-Za-z_].
2400 //
2401 if (FirstChar) {
2402 if (isalpha (Char) || (Char == '_')) {
2403 return TRUE;
2404 }
2405 } else {
2406 //
2407 // If it's not the first character, then it can
2408 // be one of [A-Za-z_0-9]
2409 //
2410 if (isalnum (Char) || (Char == '_')) {
2411 return TRUE;
2412 }
2413 }
2414
2415 return FALSE;
2416 }
2417
2418 static
2419 void
2420 RewindFile (
2421 SOURCE_FILE *SourceFile
2422 )
2423 {
2424 SourceFile->LineNum = 1;
2425 SourceFile->FileBufferPtr = SourceFile->FileBuffer;
2426 SourceFile->EndOfFile = 0;
2427 }
2428
2429 static
2430 BOOLEAN
2431 SkipTo (
2432 SOURCE_FILE *SourceFile,
2433 WCHAR WChar,
2434 BOOLEAN StopAfterNewline
2435 )
2436 {
2437 while (!EndOfFile (SourceFile)) {
2438 //
2439 // Check for the character of interest
2440 //
2441 if (SourceFile->FileBufferPtr[0] == WChar) {
2442 return TRUE;
2443 } else {
2444 if (SourceFile->FileBufferPtr[0] == UNICODE_LF) {
2445 SourceFile->LineNum++;
2446 if (StopAfterNewline) {
2447 SourceFile->FileBufferPtr++;
2448 if (SourceFile->FileBufferPtr[0] == 0) {
2449 SourceFile->FileBufferPtr++;
2450 }
2451
2452 return FALSE;
2453 }
2454 }
2455
2456 SourceFile->FileBufferPtr++;
2457 }
2458 }
2459
2460 return FALSE;
2461 }
2462
2463 static
2464 void
2465 Usage (
2466 VOID
2467 )
2468 /*++
2469
2470 Routine Description:
2471
2472 Print usage information for this utility.
2473
2474 Arguments:
2475
2476 None.
2477
2478 Returns:
2479
2480 Nothing.
2481
2482 --*/
2483 {
2484 int Index;
2485 static const char *Str[] = {
2486 "",
2487 PROGRAM_NAME " version "TOOL_VERSION " -- process unicode strings file",
2488 " Usage: "PROGRAM_NAME " -parse {parse options} [FileNames]",
2489 " "PROGRAM_NAME " -scan {scan options} [FileName]",
2490 " "PROGRAM_NAME " -dump {dump options}",
2491 " Common options include:",
2492 " -h or -? for this help information",
2493 " -db Database required name of output/input database file",
2494 " -bn BaseName for use in the .h and .c output files",
2495 " Default = "DEFAULT_BASE_NAME,
2496 " -v for verbose output",
2497 " -vdbw for verbose output when writing database",
2498 " -vdbr for verbose output when reading database",
2499 " -od FileName to specify an output database file name",
2500 " Parse options include:",
2501 " -i IncludePath add IncludePath to list of search paths",
2502 " -newdb to not read in existing database file",
2503 " -uqs to indicate that unquoted strings are used",
2504 " FileNames name of one or more unicode files to parse",
2505 " Scan options include:",
2506 " -scan scan text file(s) for STRING_TOKEN() usage",
2507 " -skipext .ext to skip scan of files with .ext filename extension",
2508 " -ignorenotfound ignore if a given STRING_TOKEN(STR) is not ",
2509 " found in the database",
2510 " FileNames one or more files to scan",
2511 " Dump options include:",
2512 " -oc FileName write string data to FileName",
2513 " -oh FileName write string defines to FileName",
2514 " -ou FileName dump database to unicode file FileName",
2515 " -lang Lang only dump for the language 'Lang'",
2516 " -if FileName to specify an indirection file",
2517 " -hpk FileName to create an HII export pack of the strings",
2518 "",
2519 " The expected process is to parse a unicode string file to create an initial",
2520 " database of string identifier names and string definitions. Then text files",
2521 " should be scanned for STRING_TOKEN() usages, and the referenced",
2522 " strings will be tagged as used in the database. After all files have been",
2523 " scanned, then the database should be dumped to create the necessary output",
2524 " files.",
2525 "",
2526 NULL
2527 };
2528 for (Index = 0; Str[Index] != NULL; Index++) {
2529 fprintf (stdout, "%s\n", Str[Index]);
2530 }
2531 }