Fossil

Changes On Branch commitWarning
Login

Changes On Branch commitWarning

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Changes In Branch commitWarning Excluding Merge-Ins

This is equivalent to a diff from 70b4f105 to d57f0a93

2012-11-01
03:44
Merge commit warning and looks_like_text() enhancements to trunk. Further changes based on these will occur on a branch. ... (check-in: 61825842 user: mistachkin tags: trunk)
2012-10-30
10:23
Fix a typo in a comment. ... (check-in: 1167d7b1 user: drh tags: trunk)
08:06
take over various stylistic adjustments ... (check-in: ab5a6b43 user: jan.nijtmans tags: use-blob_strip_bom)
02:17
Refactor commit warning functionality. Break out UTF-16 BOM detection into a new function. Style and comment fixes. ... (Closed-Leaf check-in: d57f0a93 user: mistachkin tags: commitWarning)
2012-10-29
15:57
(experiment) Allow 'c' for converting unicode/crlf files at user request, not complete yet. ... (check-in: 121394b0 user: jan.nijtmans tags: use-blob_strip_bom)
14:45
Allow unicode files for Wiki-rendering on Windows.

Trying to commit a file with an UTF-16 BOM will now give a warning, just as a file containing crlf

The --args command line option now accept UTF-16 files as well

... (check-in: 70b4f105 user: jan.nijtmans tags: trunk)
14:35
fix "possible unresolved merge conflict" warning message to show the pathname relative to the wd in stead of the root ... (check-in: 2a15d87e user: jan.nijtmans tags: trunk)
14:04
fix handling of "crlf-glob" option: UTF-16 BOM's always give a warning now ... (Closed-Leaf check-in: 9267dba4 user: jan.nijtmans tags: use-blob_strip_bom)

Changes to src/checkin.c.

880
881
882
883
884
885
886

887

888
889
890



891
892
893
894
895
896
897
898




899
900
901


902

903
904

905
906

907
908

909
910
911
912
913
914

915
916
917
918
919
920
921
922

923
924
925
926
927
928
929
880
881
882
883
884
885
886
887

888
889


890
891
892
893
894
895
896
897



898
899
900
901
902
903
904
905
906

907
908

909
910

911
912

913
914
915
916
917
918

919
920
921
922
923
924
925
926

927
928
929
930
931
932
933
934







+
-
+

-
-
+
+
+





-
-
-
+
+
+
+



+
+
-
+

-
+

-
+

-
+





-
+







-
+







  md5sum_blob(pOut, &mcksum);
  blob_appendf(pOut, "Z %b\n", &mcksum);
  if( pnFBcard ) *pnFBcard = nFBcard;
}

/*
** Issue a warning and give the user an opportunity to abandon out
** if a Unicode (UTF-16) byte-order-mark (BOM) or a \r\n line ending
** if unicode or a \r\n line ending is seen in a text file.
** is seen in a text file.
*/
static void encoding_warning(const Blob *p, int crnlOk, const char *zFilename){
  int looksLike;          /* return value of looks_like_text() */
static void commit_warning(const Blob *p, int crnlOk, const char *zFilename){
  int eType;              /* return value of looks_like_text() */
  int fUnicode;           /* return value of starts_with_utf16_bom() */
  char *zMsg;             /* Warning message */
  Blob fname;             /* Relative pathname of the file */
  static int allOk = 0;   /* Set to true to disable this routine */

  if( allOk ) return;
  looksLike = looks_like_text(p);
  if( looksLike<0 ){
    const char *type;
  eType = looks_like_text(p);
  fUnicode = starts_with_utf16_bom(p);
  if( eType==-1 || fUnicode ){
    const char *zWarning;
    Blob ans;
    char cReply;

    if( eType==-1 && fUnicode ){
      zWarning = "Unicode and CR/NL line endings";
    if( looksLike&1 ){
    }else if( eType==-1 ){
      if( crnlOk ){
        return; /* We don't want CrLf warnings for this file. */
        return; /* We don't want CR/NL warnings for this file. */
      }
      type = "CR/NL line endings";
      zWarning = "CR/NL line endings";
    }else{
      type = "unicode";
      zWarning = "Unicode";
    }
    file_relative_name(zFilename, &fname, 0);
    blob_zero(&ans);
    zMsg = mprintf(
         "%s contains %s; commit anyhow (a=all/y/N)?",
         blob_str(&fname), type);
         blob_str(&fname), zWarning);
    prompt_user(zMsg, &ans);
    fossil_free(zMsg);
    cReply = blob_str(&ans)[0];
    if( cReply=='a' || cReply=='A' ){
      allOk = 1;
    }else if( cReply!='y' && cReply!='Y' ){
      fossil_fatal("Abandoning commit due to %s in %s",
                   type, blob_str(&fname));
                   zWarning, blob_str(&fname));
    }
    blob_reset(&ans);
    blob_reset(&fname);
  }
}

/*
1230
1231
1232
1233
1234
1235
1236
1237

1238
1239
1240
1241
1242
1243
1244
1235
1236
1237
1238
1239
1240
1241

1242
1243
1244
1245
1246
1247
1248
1249







-
+







    blob_zero(&content);
    if( file_wd_islink(zFullname) ){
      /* Instead of file content, put link destination path */
      blob_read_link(&content, zFullname);
    }else{
      blob_read_from_file(&content, zFullname);
    }
    encoding_warning(&content, crnlOk, zFullname);
    commit_warning(&content, crnlOk, zFullname);
    if( chnged==1 && contains_merge_marker(&content) ){
      Blob fname; /* Relative pathname of the file */

      nConflict++;
      file_relative_name(zFullname, &fname, 0);
      fossil_print("possible unresolved merge conflict in %s\n",
                   blob_str(&fname));

Changes to src/diff.c.

46
47
48
49
50
51
52
53

54
55
56
57
58
59
60
46
47
48
49
50
51
52

53
54
55
56
57
58
59
60







-
+







*/
#define DIFF_CANNOT_COMPUTE_BINARY \
    "cannot compute difference between binary files\n"

#define DIFF_CANNOT_COMPUTE_SYMLINK \
    "cannot compute difference between symlink and regular file\n"

#define looks_like_binary(blob) ((looks_like_text(blob)&1) == 0)
#define looks_like_binary(blob) (looks_like_text((blob)) == 0)
#endif /* INTERFACE */

/*
** Maximum length of a line in a text file.  (8192)
*/
#define LENGTH_MASK_SZ  13
#define LENGTH_MASK     ((1<<LENGTH_MASK_SZ)-1)
168
169
170
171
172
173
174


175
176
177
178












179



180
181
182
183
184
185

186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205

206
207
208
209
210
211
212
213
214
215
216
217


















218
219
220
221
222
223
224
168
169
170
171
172
173
174
175
176




177
178
179
180
181
182
183
184
185
186
187
188

189
190
191
192
193
194
195
196

197
198
199
200
201
202
203







204
205
206
207
208
209

210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247







+
+
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+
+
-
+
+
+





-
+






-
-
-
-
-
-
-






-
+












+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+








  /* Return results */
  *pnLine = nLine;
  return a;
}

/*
** This function attempts to scan each logical line within the blob to
** determine the type of content it appears to contain.  Possible return
** Returns 1, if everything OK
** Returns 0 if the specified content appears to be binary or
** contains a line that is too long
** Returns -1, if the file appears text, but it contains CrLf
** values are:
**
**  (1) -- The content appears to consist entirely of text, with lines
**         delimited by line-feed characters; however, the encoding may
**         not be UTF-8.
**
**  (0) -- The content appears to be binary because it contains embedded
**         NUL (\000) characters or an extremely long line.  Since this
**         function does not understand UTF-16, it may falsely consider
**         UTF-16 text to be binary.
**
** (-1) -- The content appears to consist entirely of text, with lines
** Returns -2, if the file starts with an UTF-16 BOM (le or be)
**         delimited by carriage-return, line-feed pairs; however, the
**         encoding may not be UTF-8.
**
*/
int looks_like_text(const Blob *pContent){
  const char *z = blob_buffer(pContent);
  unsigned int n = blob_size(pContent);
  int j, c;
  int result = 1;  /* Assume text with no CrLf */
  int result = 1;  /* Assume text with no CR/NL */

  /* Check individual lines.
  */
  if( n==0 ) return result;  /* Empty file -> text */
  c = *z;
  if( c==0 ) return 0;  /* \000 byte in a file -> binary */
  if ( n > 1 ){
    if ( (c==(char)0xff) && (z[1]==(char)0xfe) ){
      return -2;
    } else if ( (c==(char)0xfe) && (z[1]==(char)0xff) ){
      return -2;
    }
  }
  j = (c!='\n');
  while( --n>0 ){
    c = *++z; ++j;
    if( c==0 ) return 0;  /* \000 byte in a file -> binary */
    if( c=='\n' ){
      if( z[-1]=='\r' ){
        result = -1;  /* Contains CrLf, continue */
        result = -1;  /* Contains CR/NL, continue */
      }
      if( j>LENGTH_MASK ){
        return 0;  /* Very long line -> binary */
      }
      j = 0;
    }
  }
  if( j>LENGTH_MASK ){
    return 0;  /* Very long line -> binary */
  }
  return result;  /* No problems seen -> not binary */
}

/*
** This function returns non-zero if the blob starts with a UTF-16le or
** UTF-16be byte-order-mark (BOM).
*/
int starts_with_utf16_bom(const Blob *pContent){
  const char *z = blob_buffer(pContent);
  int c1, c2;

  if( blob_size(pContent)<2 ) return 0;
  c1 = z[0]; c2 = z[1];
  if( (c1==(char)0xff) && (c2==(char)0xfe) ){
    return 1;
  }else if( (c1==(char)0xff) && (c2==(char)0xfe) ){
    return 1;
  }
  return 0;
}

/*
** Return true if two DLine elements are identical.
*/
static int same_dline(DLine *pA, DLine *pB){
  return pA->h==pB->h && memcmp(pA->z,pB->z,pA->h & LENGTH_MASK)==0;
}