Fossil

Check-in [2faa1a27]
Login

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

Overview
Comment:Fix indenting behavior of --ignore-space-at-sol option
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | diff-eolws
Files: files | file ages | folders
SHA1:2faa1a272bf2b5934dc9832e6edfe3548db06bdf
User & Date: jan.nijtmans 2014-03-03 11:17:54
Context
2014-03-03
11:37
Oops check-in: 65397b57 user: jan.nijtmans tags: diff-eolws
11:17
Fix indenting behavior of --ignore-space-at-sol option check-in: 2faa1a27 user: jan.nijtmans tags: diff-eolws
10:50
Implement new --ignore-space-at-sol option for all diff variants. And "-w" which means (for now) both --ignore-space-at-sol and --ignore-space-at-eol.

Problem: diffs using --ignore-space-at-sol don't indent any more. to be solved. check-in: 6392c032 user: jan.nijtmans tags: diff-eolws

Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to src/diff.c.

53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
..
75
76
77
78
79
80
81

82
83
84
85
86
87
88
...
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
...
163
164
165
166
167
168
169

170
171
172
173
174




175

176

177
178
179
180
181
182
183
...
229
230
231
232
233
234
235



236
237
238
239
240



241
242
243
244
245
246
247
....
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
....
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808

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

#define DIFF_TOO_MANY_CHANGES \
    "more than 10,000 changes\n"

#define DIFF_EOLWS_ONLY \
    "only end-of-line whitespace changed\n"

/*
** Maximum length of a line in a text file, in bytes.  (2**13 = 8192 bytes)
*/
#define LENGTH_MASK_SZ  13
#define LENGTH_MASK     ((1<<LENGTH_MASK_SZ)-1)

................................................................................
** of the line.  If any line is longer than LENGTH_MASK characters,
** the file is considered binary.
*/
typedef struct DLine DLine;
struct DLine {
  const char *z;        /* The text of the line */
  unsigned int h;       /* Hash of the line */

  unsigned int iNext;   /* 1+(Index of next line with same the same hash) */

  /* an array of DLine elements serves two purposes.  The fields
  ** above are one per line of input text.  But each entry is also
  ** a bucket in a hash table, as follows: */
  unsigned int iHash;   /* 1+(first entry in the hash chain) */
};
................................................................................
** Return 0 if the file is binary or contains a line that is
** too long.
**
** Profiling show that in most cases this routine consumes the bulk of
** the CPU time on a diff.
*/
static DLine *break_into_lines(const char *z, int n, int *pnLine, u64 diffFlags){
  int nLine, i, j, k, s, x;
  unsigned int h, h2;
  DLine *a;

  /* Count the number of lines.  Allocate space to hold
  ** the returned array.
  */
  for(i=j=0, nLine=1; i<n; i++, j++){
................................................................................
  }

  /* Fill in the array */
  for(i=0; i<nLine; i++){
    for(j=0; z[j] && z[j]!='\n'; j++){}
    k = j;
    s = 0;

    if( diffFlags & DIFF_IGNORE_EOLWS ){
      while( k>0 && fossil_isspace(z[k-1]) ){ k--; }
    }
    if( diffFlags & DIFF_IGNORE_SOLWS ){
      while( s<k && fossil_isspace(z[s]) ){ s++; }




    }

    a[i].z = z+s;

    for(h=0, x=s; x<k; x++){
      h = h ^ (h<<2) ^ z[x];
    }
    a[i].h = h = (h<<LENGTH_MASK_SZ) | (k-s);
    h2 = h % nLine;
    a[i].iNext = a[h2].iHash;
    a[h2].iHash = i+1;
................................................................................
    if( pRe && re_dline_match(pRe, pLine, 1)==0 ){
      cPrefix = ' ';
    }else if( cPrefix=='+' ){
      blob_append(pOut, "<span class=\"diffadd\">", -1);
    }else if( cPrefix=='-' ){
      blob_append(pOut, "<span class=\"diffrm\">", -1);
    }



    htmlize_to_blob(pOut, pLine->z, (pLine->h & LENGTH_MASK));
    if( cPrefix!=' ' ){
      blob_append(pOut, "</span>", -1);
    }
  }else{



    blob_append(pOut, pLine->z, pLine->h & LENGTH_MASK);
  }
  blob_append(pOut, "\n", 1);
}

/*
** Add two line numbers to the beginning of an output line for a context
................................................................................
int *text_diff(
  Blob *pA_Blob,   /* FROM file */
  Blob *pB_Blob,   /* TO file */
  Blob *pOut,      /* Write diff here if not NULL */
  ReCompiled *pRe, /* Only output changes where this Regexp matches */
  u64 diffFlags    /* DIFF_* flags defined above */
){
  int ignoreEolWs; /* Ignore whitespace at the end of lines */
  DContext c;

  if( diffFlags & DIFF_INVERT ){
    Blob *pTemp = pA_Blob;
    pA_Blob = pB_Blob;
    pB_Blob = pTemp;
  }
  ignoreEolWs = (diffFlags & DIFF_IGNORE_EOLWS)!=0;
  blob_to_utf8_no_bom(pA_Blob, 0);
  blob_to_utf8_no_bom(pB_Blob, 0);

  /* Prepare the input files */
  memset(&c, 0, sizeof(c));
  c.aFrom = break_into_lines(blob_str(pA_Blob), blob_size(pA_Blob),
                             &c.nFrom, diffFlags);
................................................................................
      diff_errmsg(pOut, DIFF_CANNOT_COMPUTE_BINARY, diffFlags);
    }
    return 0;
  }

  /* Compute the difference */
  diff_all(&c);
  if( ignoreEolWs && c.nEdit==6 && c.aEdit[1]==0 && c.aEdit[2]==0 ){
    fossil_free(c.aFrom);
    fossil_free(c.aTo);
    fossil_free(c.aEdit);
    if( pOut ) diff_errmsg(pOut, DIFF_EOLWS_ONLY, diffFlags);
    return 0;
  }
  if( (diffFlags & DIFF_NOTTOOBIG)!=0 ){
    int i, m, n;
    int *a = c.aEdit;
    int mx = c.nEdit;
    for(i=m=n=0; i<mx; i+=3){ m += a[i]; n += a[i+1]+a[i+2]; }







|
|







 







>







 







|







 







>





>
>
>
>
|
>

>







 







>
>
>





>
>
>







 







|







|







 







|



|







53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
..
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
...
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
...
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
...
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
....
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
....
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822

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

#define DIFF_TOO_MANY_CHANGES \
    "more than 10,000 changes\n"

#define DIFF_WHITESPACE_ONLY \
    "whitespace changes only\n"

/*
** Maximum length of a line in a text file, in bytes.  (2**13 = 8192 bytes)
*/
#define LENGTH_MASK_SZ  13
#define LENGTH_MASK     ((1<<LENGTH_MASK_SZ)-1)

................................................................................
** of the line.  If any line is longer than LENGTH_MASK characters,
** the file is considered binary.
*/
typedef struct DLine DLine;
struct DLine {
  const char *z;        /* The text of the line */
  unsigned int h;       /* Hash of the line */
  unsigned int indent;  /* Indent of the line. Only !=0 with --ignore-space-at sol option */
  unsigned int iNext;   /* 1+(Index of next line with same the same hash) */

  /* an array of DLine elements serves two purposes.  The fields
  ** above are one per line of input text.  But each entry is also
  ** a bucket in a hash table, as follows: */
  unsigned int iHash;   /* 1+(first entry in the hash chain) */
};
................................................................................
** Return 0 if the file is binary or contains a line that is
** too long.
**
** Profiling show that in most cases this routine consumes the bulk of
** the CPU time on a diff.
*/
static DLine *break_into_lines(const char *z, int n, int *pnLine, u64 diffFlags){
  int nLine, i, j, k, s, indent, x;
  unsigned int h, h2;
  DLine *a;

  /* Count the number of lines.  Allocate space to hold
  ** the returned array.
  */
  for(i=j=0, nLine=1; i<n; i++, j++){
................................................................................
  }

  /* Fill in the array */
  for(i=0; i<nLine; i++){
    for(j=0; z[j] && z[j]!='\n'; j++){}
    k = j;
    s = 0;
    indent = 0;
    if( diffFlags & DIFF_IGNORE_EOLWS ){
      while( k>0 && fossil_isspace(z[k-1]) ){ k--; }
    }
    if( diffFlags & DIFF_IGNORE_SOLWS ){
      while( s<k && fossil_isspace(z[s]) ){ s++; }
      if( z[s]=='\t' ){
        indent = ((indent+9)/8)*8;
      }else if( z[s]==' ' ){
        indent++;
      }
    }
    a[i].z = z+s;
    a[i].indent = s;
    for(h=0, x=s; x<k; x++){
      h = h ^ (h<<2) ^ z[x];
    }
    a[i].h = h = (h<<LENGTH_MASK_SZ) | (k-s);
    h2 = h % nLine;
    a[i].iNext = a[h2].iHash;
    a[h2].iHash = i+1;
................................................................................
    if( pRe && re_dline_match(pRe, pLine, 1)==0 ){
      cPrefix = ' ';
    }else if( cPrefix=='+' ){
      blob_append(pOut, "<span class=\"diffadd\">", -1);
    }else if( cPrefix=='-' ){
      blob_append(pOut, "<span class=\"diffrm\">", -1);
    }
    if( pLine->indent ){
      blob_appendf(pOut, "%*s", pLine->indent, " ");
    }
    htmlize_to_blob(pOut, pLine->z, (pLine->h & LENGTH_MASK));
    if( cPrefix!=' ' ){
      blob_append(pOut, "</span>", -1);
    }
  }else{
    if( pLine->indent ){
      blob_appendf(pOut, "%*s", pLine->indent, " ");
    }
    blob_append(pOut, pLine->z, pLine->h & LENGTH_MASK);
  }
  blob_append(pOut, "\n", 1);
}

/*
** Add two line numbers to the beginning of an output line for a context
................................................................................
int *text_diff(
  Blob *pA_Blob,   /* FROM file */
  Blob *pB_Blob,   /* TO file */
  Blob *pOut,      /* Write diff here if not NULL */
  ReCompiled *pRe, /* Only output changes where this Regexp matches */
  u64 diffFlags    /* DIFF_* flags defined above */
){
  int ignoreWs; /* Ignore whitespace */
  DContext c;

  if( diffFlags & DIFF_INVERT ){
    Blob *pTemp = pA_Blob;
    pA_Blob = pB_Blob;
    pB_Blob = pTemp;
  }
  ignoreWs = (diffFlags & (DIFF_IGNORE_SOLWS|DIFF_IGNORE_EOLWS))!=0;
  blob_to_utf8_no_bom(pA_Blob, 0);
  blob_to_utf8_no_bom(pB_Blob, 0);

  /* Prepare the input files */
  memset(&c, 0, sizeof(c));
  c.aFrom = break_into_lines(blob_str(pA_Blob), blob_size(pA_Blob),
                             &c.nFrom, diffFlags);
................................................................................
      diff_errmsg(pOut, DIFF_CANNOT_COMPUTE_BINARY, diffFlags);
    }
    return 0;
  }

  /* Compute the difference */
  diff_all(&c);
  if( ignoreWs && c.nEdit==6 && c.aEdit[1]==0 && c.aEdit[2]==0 ){
    fossil_free(c.aFrom);
    fossil_free(c.aTo);
    fossil_free(c.aEdit);
    if( pOut ) diff_errmsg(pOut, DIFF_WHITESPACE_ONLY, diffFlags);
    return 0;
  }
  if( (diffFlags & DIFF_NOTTOOBIG)!=0 ){
    int i, m, n;
    int *a = c.aEdit;
    int mx = c.nEdit;
    for(i=m=n=0; i<mx; i+=3){ m += a[i]; n += a[i+1]+a[i+2]; }