blame.h

   1 #ifndef BLAME_H
   2 #define BLAME_H
   3
   4 #include "cache.h"
   5 #include "commit.h"
   6 #include "xdiff-interface.h"
   7 #include "revision.h"
   8 #include "prio-queue.h"
   9 #include "diff.h"
  10
  11 #define PICKAXE_BLAME_MOVE              01
  12 #define PICKAXE_BLAME_COPY              02
  13 #define PICKAXE_BLAME_COPY_HARDER       04
  14 #define PICKAXE_BLAME_COPY_HARDEST      010
  15
  16 #define BLAME_DEFAULT_MOVE_SCORE        20
  17 #define BLAME_DEFAULT_COPY_SCORE        40
  18
  19 /*
  20  * One blob in a commit that is being suspected
  21  */
  22 struct blame_origin {
  23         int refcnt;
  24         /* Record preceding blame record for this blob */
  25         struct blame_origin *previous;
  26         /* origins are put in a list linked via `next' hanging off the
  27          * corresponding commit's util field in order to make finding
  28          * them fast.  The presence in this chain does not count
  29          * towards the origin's reference count.  It is tempting to
  30          * let it count as long as the commit is pending examination,
  31          * but even under circumstances where the commit will be
  32          * present multiple times in the priority queue of unexamined
  33          * commits, processing the first instance will not leave any
  34          * work requiring the origin data for the second instance.  An
  35          * interspersed commit changing that would have to be
  36          * preexisting with a different ancestry and with the same
  37          * commit date in order to wedge itself between two instances
  38          * of the same commit in the priority queue _and_ produce
  39          * blame entries relevant for it.  While we don't want to let
  40          * us get tripped up by this case, it certainly does not seem
  41          * worth optimizing for.
  42          */
  43         struct blame_origin *next;
  44         struct commit *commit;
  45         /* `suspects' contains blame entries that may be attributed to
  46          * this origin's commit or to parent commits.  When a commit
  47          * is being processed, all suspects will be moved, either by
  48          * assigning them to an origin in a different commit, or by
  49          * shipping them to the scoreboard's ent list because they
  50          * cannot be attributed to a different commit.
  51          */
  52         struct blame_entry *suspects;
  53         mmfile_t file;
  54         int num_lines;
  55         void *fingerprints;
  56         struct object_id blob_oid;
  57         unsigned short mode;
  58         /* guilty gets set when shipping any suspects to the final
  59          * blame list instead of other commits
  60          */
  61         char guilty;
  62         char path[FLEX_ARRAY];
  63 };
  64
  65 /*
  66  * Each group of lines is described by a blame_entry; it can be split
  67  * as we pass blame to the parents.  They are arranged in linked lists
  68  * kept as `suspects' of some unprocessed origin, or entered (when the
  69  * blame origin has been finalized) into the scoreboard structure.
  70  * While the scoreboard structure is only sorted at the end of
  71  * processing (according to final image line number), the lists
  72  * attached to an origin are sorted by the target line number.
  73  */
  74 struct blame_entry {
  75         struct blame_entry *next;
  76
  77         /* the first line of this group in the final image;
  78          * internally all line numbers are 0 based.
  79          */
  80         int lno;
  81
  82         /* how many lines this group has */
  83         int num_lines;
  84
  85         /* the commit that introduced this group into the final image */
  86         struct blame_origin *suspect;
  87
  88         /* the line number of the first line of this group in the
  89          * suspect's file; internally all line numbers are 0 based.
  90          */
  91         int s_lno;
  92
  93         /* how significant this entry is -- cached to avoid
  94          * scanning the lines over and over.
  95          */
  96         unsigned score;
  97         int ignored;
  98         int unblamable;
  99 };
 100
 101 /*
 102  * The current state of the blame assignment.
 103  */
 104 struct blame_scoreboard {
 105         /* the final commit (i.e. where we started digging from) */
 106         struct commit *final;
 107         /* Priority queue for commits with unassigned blame records */
 108         struct prio_queue commits;
 109         struct repository *repo;
 110         struct rev_info *revs;
 111         const char *path;
 112
 113         /*
 114          * The contents in the final image.
 115          * Used by many functions to obtain contents of the nth line,
 116          * indexed with scoreboard.lineno[blame_entry.lno].
 117          */
 118         const char *final_buf;
 119         unsigned long final_buf_size;
 120
 121         /* linked list of blames */
 122         struct blame_entry *ent;
 123
 124         struct oidset ignore_list;
 125
 126         /* look-up a line in the final buffer */
 127         int num_lines;
 128         int *lineno;
 129
 130         /* stats */
 131         int num_read_blob;
 132         int num_get_patch;
 133         int num_commits;
 134
 135         /*
 136          * blame for a blame_entry with score lower than these thresholds
 137          * is not passed to the parent using move/copy logic.
 138          */
 139         unsigned move_score;
 140         unsigned copy_score;
 141
 142         /* use this file's contents as the final image */
 143         const char *contents_from;
 144
 145         /* flags */
 146         int reverse;
 147         int show_root;
 148         int xdl_opts;
 149         int no_whole_file_rename;
 150         int debug;
 151
 152         /* callbacks */
 153         void(*on_sanity_fail)(struct blame_scoreboard *, int);
 154         void(*found_guilty_entry)(struct blame_entry *, void *);
 155
 156         void *found_guilty_entry_data;
 157 };
 158
 159 /*
 160  * Origin is refcounted and usually we keep the blob contents to be
 161  * reused.
 162  */
 163 static inline struct blame_origin *blame_origin_incref(struct blame_origin *o)
 164 {
 165         if (o)
 166                 o->refcnt++;
 167         return o;
 168 }
 169 void blame_origin_decref(struct blame_origin *o);
 170
 171 void blame_coalesce(struct blame_scoreboard *sb);
 172 void blame_sort_final(struct blame_scoreboard *sb);
 173 unsigned blame_entry_score(struct blame_scoreboard *sb, struct blame_entry *e);
 174 void assign_blame(struct blame_scoreboard *sb, int opt);
 175 const char *blame_nth_line(struct blame_scoreboard *sb, long lno);
 176
 177 void init_scoreboard(struct blame_scoreboard *sb);
 178 void setup_scoreboard(struct blame_scoreboard *sb,
 179                       const char *path,
 180                       struct blame_origin **orig);
 181
 182 struct blame_entry *blame_entry_prepend(struct blame_entry *head,
 183                                         long start, long end,
 184                                         struct blame_origin *o);
 185
 186 struct blame_origin *get_blame_suspects(struct commit *commit);
 187
 188 #endif /* BLAME_H */