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         struct object_id blob_oid;
  55         unsigned mode;
  56         /* guilty gets set when shipping any suspects to the final
  57          * blame list instead of other commits
  58          */
  59         char guilty;
  60         char path[FLEX_ARRAY];
  61 };
  62
  63 /*
  64  * Each group of lines is described by a blame_entry; it can be split
  65  * as we pass blame to the parents.  They are arranged in linked lists
  66  * kept as `suspects' of some unprocessed origin, or entered (when the
  67  * blame origin has been finalized) into the scoreboard structure.
  68  * While the scoreboard structure is only sorted at the end of
  69  * processing (according to final image line number), the lists
  70  * attached to an origin are sorted by the target line number.
  71  */
  72 struct blame_entry {
  73         struct blame_entry *next;
  74
  75         /* the first line of this group in the final image;
  76          * internally all line numbers are 0 based.
  77          */
  78         int lno;
  79
  80         /* how many lines this group has */
  81         int num_lines;
  82
  83         /* the commit that introduced this group into the final image */
  84         struct blame_origin *suspect;
  85
  86         /* the line number of the first line of this group in the
  87          * suspect's file; internally all line numbers are 0 based.
  88          */
  89         int s_lno;
  90
  91         /* how significant this entry is -- cached to avoid
  92          * scanning the lines over and over.
  93          */
  94         unsigned score;
  95 };
  96
  97 /*
  98  * The current state of the blame assignment.
  99  */
 100 struct blame_scoreboard {
 101         /* the final commit (i.e. where we started digging from) */
 102         struct commit *final;
 103         /* Priority queue for commits with unassigned blame records */
 104         struct prio_queue commits;
 105         struct repository *repo;
 106         struct rev_info *revs;
 107         const char *path;
 108
 109         /*
 110          * The contents in the final image.
 111          * Used by many functions to obtain contents of the nth line,
 112          * indexed with scoreboard.lineno[blame_entry.lno].
 113          */
 114         const char *final_buf;
 115         unsigned long final_buf_size;
 116
 117         /* linked list of blames */
 118         struct blame_entry *ent;
 119
 120         /* look-up a line in the final buffer */
 121         int num_lines;
 122         int *lineno;
 123
 124         /* stats */
 125         int num_read_blob;
 126         int num_get_patch;
 127         int num_commits;
 128
 129         /*
 130          * blame for a blame_entry with score lower than these thresholds
 131          * is not passed to the parent using move/copy logic.
 132          */
 133         unsigned move_score;
 134         unsigned copy_score;
 135
 136         /* use this file's contents as the final image */
 137         const char *contents_from;
 138
 139         /* flags */
 140         int reverse;
 141         int show_root;
 142         int xdl_opts;
 143         int no_whole_file_rename;
 144         int debug;
 145
 146         /* callbacks */
 147         void(*on_sanity_fail)(struct blame_scoreboard *, int);
 148         void(*found_guilty_entry)(struct blame_entry *, void *);
 149
 150         void *found_guilty_entry_data;
 151 };
 152
 153 /*
 154  * Origin is refcounted and usually we keep the blob contents to be
 155  * reused.
 156  */
 157 static inline struct blame_origin *blame_origin_incref(struct blame_origin *o)
 158 {
 159         if (o)
 160                 o->refcnt++;
 161         return o;
 162 }
 163 void blame_origin_decref(struct blame_origin *o);
 164
 165 void blame_coalesce(struct blame_scoreboard *sb);
 166 void blame_sort_final(struct blame_scoreboard *sb);
 167 unsigned blame_entry_score(struct blame_scoreboard *sb, struct blame_entry *e);
 168 void assign_blame(struct blame_scoreboard *sb, int opt);
 169 const char *blame_nth_line(struct blame_scoreboard *sb, long lno);
 170
 171 void init_scoreboard(struct blame_scoreboard *sb);
 172 void setup_scoreboard(struct blame_scoreboard *sb,
 173                       const char *path,
 174                       struct blame_origin **orig);
 175
 176 struct blame_entry *blame_entry_prepend(struct blame_entry *head,
 177                                         long start, long end,
 178                                         struct blame_origin *o);
 179
 180 extern struct blame_origin *get_blame_suspects(struct commit *commit);
 181
 182 #endif /* BLAME_H */