What's cooking (2017/09 #01)
Junio C Hamano committed
Sep 6, 2017 at 13:36 UTC
f575a57c0113603097b386e169f1335f1b96abaf
1 file changed
+152
-363
whats-cooking.txt
+152
-363
@@ -1,10 +1,10 @@
1
To: git@vger.kernel.org
2
Bcc: lwn@lwn.net
3
-Subject: What's cooking in git.git (Aug 2017, #07; Sat, 26)
4
-X-master-at: 238e487ea943f80734cc6dad665e7238b8cbc7ff
5
-X-next-at: f28d330327670643ea4cc6f8feceac4024df00a0
3
+Subject: What's cooking in git.git (Sep 2017, #01; Wed, 6)
4
+X-master-at: 3ec7d702a89c647ddf42a59bc3539361367de9d5
5
+X-next-at: 1a2e63c10015eb714409797caf56c3d4ae975672
6
7
-What's cooking in git.git (Aug 2017, #07; Sat, 26)
7
+What's cooking in git.git (Sep 2017, #01; Wed, 6)
8
--------------------------------------------------
9
10
Here are the topics that have been cooking. Commits prefixed with
@@ -12,12 +12,9 @@ Here are the topics that have been cooking. Commits prefixed with
12
'+' are in 'next'. The ones marked with '.' do not appear in any of
13
the integration branches, but I am still holding onto them.
14
15
-As I am preparing to go offline next week for about a week, a
16
-handful of topics have been merged to 'master' and 'next' is getting
17
-thinner. There are a few topics of importance that are still only
18
-on the mailing list archive and not in my tree that I'd like to
19
-review them and help them get into a reasonable shape, but that
20
-hasn't happened yet.
15
+We are at week #5 of this cycle. It seems that people had a
16
+productive week while I was away, which I am reasonably happy about
17
+;-)
18
19
You can find the changes described here in the integration branches
20
of the repositories listed at
@@ -27,348 +24,94 @@ of the repositories listed at
24
--------------------------------------------------
25
[Graduated to "master"]
26
30
-* bc/vcs-svn-cleanup (2017-08-20) 2 commits
31
- (merged to 'next' on 2017-08-22 at d8494b5d5b)
32
- + vcs-svn: rename repo functions to "svn_repo"
33
- + vcs-svn: remove unused prototypes
34
- (this branch is used by bc/hash-algo and jn/vcs-svn-cleanup.)
35
-
36
- Code clean-up.
37
-
38
-
39
-* bw/submodule-config-cleanup (2017-08-03) 17 commits
40
- (merged to 'next' on 2017-08-24 at e35d58c4c6)
41
- + submodule: remove gitmodules_config
42
- + unpack-trees: improve loading of .gitmodules
43
- + submodule-config: lazy-load a repository's .gitmodules file
44
- + submodule-config: move submodule-config functions to submodule-config.c
45
- + submodule-config: remove support for overlaying repository config
46
- + diff: stop allowing diff to have submodules configured in .git/config
47
- + submodule: remove submodule_config callback routine
48
- + unpack-trees: don't respect submodule.update
49
- + submodule: don't rely on overlayed config when setting diffopts
50
- + fetch: don't overlay config with submodule-config
51
- + submodule--helper: don't overlay config in update-clone
52
- + submodule--helper: don't overlay config in remote_submodule_branch
53
- + add, reset: ensure submodules can be added or reset
54
- + submodule: don't use submodule_from_name
55
- + t7411: check configuration parsing errors
56
- + Merge branch 'bc/object-id' into bw/submodule-config-cleanup
57
- + Merge branch 'bw/grep-recurse-submodules' into bw/submodule-config-cleanup
58
-
59
- Code clean-up to avoid mixing values read from the .gitmodules file
60
- and values read from the .git/config file.
61
- So, after the recent discussion around submodule.<name>.update (and
62
- its resolution "use submodule.<name>.active; "reset --hard" must
63
- ignore the .update thing"), this is now good to go, I presume?
64
- Please yell at me that I am clueless if that is not the case ;-)
65
-
66
-
67
-* jc/cutoff-config (2017-08-22) 6 commits
68
- (merged to 'next' on 2017-08-24 at 2dcbf9ae04)
69
- + rerere: allow approxidate in gc.rerereResolved/gc.rerereUnresolved
70
- + rerere: represent time duration in timestamp_t internally
71
- + t4200: parameterize "rerere gc" custom expiry test
72
- + t4200: gather "rerere gc" together
73
- + t4200: make "rerere gc" test more robust
74
- + t4200: give us a clean slate after "rerere gc" tests
75
-
76
- "[gc] rerereResolved = 5.days" used to be invalid, as the variable
77
- is defined to take an integer counting the number of days. It now
78
- is allowed.
79
-
80
-
81
-* jk/trailers-parse (2017-08-20) 9 commits
82
- (merged to 'next' on 2017-08-22 at 2d4d937141)
83
- + doc/interpret-trailers: fix "the this" typo
84
- (merged to 'next' on 2017-08-19 at faa63f8196)
85
- + pretty: support normalization options for %(trailers)
86
- + t4205: refactor %(trailers) tests
87
- + pretty: move trailer formatting to trailer.c
88
- + interpret-trailers: add --parse convenience option
89
- + interpret-trailers: add an option to unfold values
90
- + interpret-trailers: add an option to show only existing trailers
91
- + interpret-trailers: add an option to show only the trailers
92
- + trailer: put process_trailers() options into a struct
93
-
94
- "git interpret-trailers" has been taught a "--parse" and a few
95
- other options to make it easier for scripts to grab existing
96
- trailer lines from a commit log message.
97
- Possible leftover bits are mentioned in
98
- cf. <20170820094009.z23wclpku35txflg@sigill.intra.peff.net>
99
-
100
-
101
-* jn/vcs-svn-cleanup (2017-08-23) 4 commits
102
- (merged to 'next' on 2017-08-24 at c184f08f5f)
103
- + vcs-svn: move remaining repo_tree functions to fast_export.h
104
- + vcs-svn: remove repo_delete wrapper function
105
- + vcs-svn: remove custom mode constants
106
- + vcs-svn: remove more unused prototypes and declarations
107
- (this branch uses bc/vcs-svn-cleanup; is tangled with bc/hash-algo.)
108
-
109
- Code clean-up.
110
-
111
-
112
-* js/gitweb-raw-blob-link-in-history (2017-08-22) 1 commit
113
- (merged to 'next' on 2017-08-23 at 16258266e2)
114
- + gitweb: add 'raw' blob_plain link in history overview
115
-
116
- "gitweb" shows a link to visit the 'raw' contents of blbos in the
117
- history overview page.
27
+* ks/branch-set-upstream (2017-08-17) 3 commits
28
+ (merged to 'next' on 2017-08-22 at 10fd938390)
29
+ + branch: quote branch/ref names to improve readability
30
+ + builtin/branch: stop supporting the "--set-upstream" option
31
+ + t3200: cleanup cruft of a test
32
33
+ "branch --set-upstream" that has been deprecated in Git 1.8 has
34
+ finally been retired.
35
120
-* jt/diff-color-move-fix (2017-08-16) 3 commits
121
- (merged to 'next' on 2017-08-19 at 6ae3831c8c)
122
- + diff: define block by number of alphanumeric chars
123
- + diff: respect MIN_BLOCK_LENGTH for last block
124
- + diff: avoid redundantly clearing a flag
125
- (this branch uses sb/diff-color-move.)
36
127
- A handful of bugfixes and an improvement to "diff --color-moved".
37
+* po/read-graft-line (2017-08-18) 4 commits
38
+ (merged to 'next' on 2017-08-22 at 1e3fe0d3a1)
39
+ + commit: rewrite read_graft_line
40
+ + commit: allocate array using object_id size
41
+ + commit: replace the raw buffer with strbuf in read_graft_line
42
+ + sha1_file: fix definition of null_sha1
43
44
+ Conversion from uchar[20] to struct object_id continues; this is to
45
+ ensure that we do not assume sizeof(struct object_id) is the same
46
+ as the length of SHA-1 hash (or length of longest hash we support).
47
130
-* jt/doc-pack-objects-fix (2017-08-23) 1 commit
131
- (merged to 'next' on 2017-08-24 at 7e4c3c0c9f)
132
- + Doc: clarify that pack-objects makes packs, plural
48
134
- Doc updates.
49
+* rs/archive-excluded-directory (2017-08-19) 3 commits
50
+ (merged to 'next' on 2017-08-22 at 1853597c35)
51
+ + archive: don't queue excluded directories
52
+ + archive: factor out helper functions for handling attributes
53
+ + t5001: add tests for export-ignore attributes and exclude pathspecs
54
55
+ "git archive" did not work well with pathspecs and the
56
+ export-ignore attribute.
57
137
-* jt/packmigrate (2017-08-23) 23 commits
138
- (merged to 'next' on 2017-08-24 at 34a628ae6c)
139
- + pack: move for_each_packed_object()
140
- + pack: move has_pack_index()
141
- + pack: move has_sha1_pack()
142
- + pack: move find_pack_entry() and make it global
143
- + pack: move find_sha1_pack()
144
- + pack: move find_pack_entry_one(), is_pack_valid()
145
- + pack: move check_pack_index_ptr(), nth_packed_object_offset()
146
- + pack: move nth_packed_object_{sha1,oid}
147
- + pack: move clear_delta_base_cache(), packed_object_info(), unpack_entry()
148
- + pack: move unpack_object_header()
149
- + pack: move get_size_from_delta()
150
- + pack: move unpack_object_header_buffer()
151
- + pack: move {,re}prepare_packed_git and approximate_object_count
152
- + pack: move install_packed_git()
153
- + pack: move add_packed_git()
154
- + pack: move unuse_pack()
155
- + pack: move use_pack()
156
- + pack: move pack-closing functions
157
- + pack: move release_pack_memory()
158
- + pack: move open_pack_index(), parse_pack_index()
159
- + pack: move pack_report()
160
- + pack: move static state variables
161
- + pack: move pack name-related functions
162
-
163
- Code movement to make it easier to hack later.
164
-
165
-
166
-* jt/stash-tests (2017-08-19) 3 commits
167
- (merged to 'next' on 2017-08-22 at 0b022697aa)
168
- + stash: add a test for stashing in a detached state
169
- + stash: add a test for when apply fails during stash branch
170
- + stash: add a test for stash create with no files
171
-
172
- Test update to improve coverage for "git stash" operations.
173
-
174
-
175
-* kw/write-index-reduce-alloc (2017-08-21) 3 commits
176
- (merged to 'next' on 2017-08-24 at 29f7daad85)
177
- + read-cache: avoid allocating every ondisk entry when writing
178
- + read-cache: fix memory leak in do_write_index
179
- + perf: add test for writing the index
180
-
181
- We used to spend more than necessary cycles allocating and freeing
182
- piece of memory while writing each index entry out. This has been
183
- optimized.
184
-
185
-
186
-* mg/killed-merge (2017-08-23) 4 commits
187
- (merged to 'next' on 2017-08-24 at 5db37fc192)
188
- + merge: save merge state earlier
189
- + merge: split write_merge_state in two
190
- + merge: clarify call chain
191
- + Documentation/git-merge: explain --continue
192
-
193
- Killing "git merge --edit" before the editor returns control left
194
- the repository in a state with MERGE_MSG but without MERGE_HEAD,
195
- which incorrectly tells the subsequent "git commit" that there was
196
- a squash merge in progress. This has been fixed.
197
-
198
-
199
-* mh/ref-lock-entry (2017-08-23) 1 commit
200
- (merged to 'next' on 2017-08-24 at 67760267f4)
201
- + refs: retry acquiring reference locks for 100ms
202
-
203
- The code to acquire a lock on a reference (e.g. while accepting a
204
- push from a client) used to immediately fail when the reference is
205
- already locked---now it waits for a very short while and retries,
206
- which can make it succeed if the lock holder was holding it during
207
- a read-only operation.
208
-
209
-
210
-* pb/trailers-from-command-line (2017-08-14) 4 commits
211
- (merged to 'next' on 2017-08-19 at b12342e49e)
212
- + interpret-trailers: fix documentation typo
213
- + interpret-trailers: add options for actions
214
- + trailers: introduce struct new_trailer_item
215
- + trailers: export action enums and corresponding lookup functions
216
-
217
- "git interpret-trailers" learned to take the trailer specifications
218
- from the command line that overrides the configured values.
219
-
220
-
221
-* po/object-id (2017-08-20) 6 commits
222
- (merged to 'next' on 2017-08-23 at 0faff383a8)
223
- + sha1_file: convert index_stream to struct object_id
224
- + sha1_file: convert hash_sha1_file_literally to struct object_id
225
- + sha1_file: convert index_fd to struct object_id
226
- + sha1_file: convert index_path to struct object_id
227
- + read-cache: convert to struct object_id
228
- + builtin/hash-object: convert to struct object_id
229
-
230
- Conversion from uchar[20] to struct object_id continues.
231
-
232
-
233
-* sb/diff-color-move (2017-06-30) 26 commits
234
- (merged to 'next' on 2017-08-19 at a239d031bc)
235
- + diff: document the new --color-moved setting
236
- + diff.c: add dimming to moved line detection
237
- + diff.c: color moved lines differently, plain mode
238
- + diff.c: color moved lines differently
239
- + diff.c: buffer all output if asked to
240
- + diff.c: emit_diff_symbol learns about DIFF_SYMBOL_SUMMARY
241
- + diff.c: emit_diff_symbol learns about DIFF_SYMBOL_STAT_SEP
242
- + diff.c: convert word diffing to use emit_diff_symbol
243
- + diff.c: convert show_stats to use emit_diff_symbol
244
- + diff.c: convert emit_binary_diff_body to use emit_diff_symbol
245
- + submodule.c: migrate diff output to use emit_diff_symbol
246
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_REWRITE_DIFF
247
- + diff.c: emit_diff_symbol learns about DIFF_SYMBOL_BINARY_FILES
248
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_HEADER
249
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_FILEPAIR_{PLUS, MINUS}
250
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_CONTEXT_INCOMPLETE
251
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_WORDS[_PORCELAIN]
252
- + diff.c: migrate emit_line_checked to use emit_diff_symbol
253
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_NO_LF_EOF
254
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_CONTEXT_FRAGINFO
255
- + diff.c: emit_diff_symbol learns DIFF_SYMBOL_CONTEXT_MARKER
256
- + diff.c: introduce emit_diff_symbol
257
- + diff.c: factor out diff_flush_patch_all_file_pairs
258
- + diff.c: move line ending check into emit_hunk_header
259
- + diff.c: readability fix
260
- + Merge branch 'sb/hashmap-customize-comparison' into sb/diff-color-move
261
- (this branch is used by jt/diff-color-move-fix.)
262
-
263
- "git diff" has been taught to optionally paint new lines that are
264
- the same as deleted lines elsewhere differently from genuinely new
265
- lines.
266
-
267
-
268
-* tb/apply-with-crlf (2017-08-19) 2 commits
269
- (merged to 'next' on 2017-08-22 at f92bf67061)
270
- + apply: file commited with CRLF should roundtrip diff and apply
271
- + convert: add SAFE_CRLF_KEEP_CRLF
272
- (this branch is tangled with jc/apply-with-crlf.)
273
-
274
- "git apply" that is used as a better "patch -p1" failed to apply a
275
- taken from a file with CRLF line endings to a file with CRLF line
276
- endings. The root cause was because it misused convert_to_git()
277
- that tried to do "safe-crlf" processing by looking at the index
278
- entry at the same path, which is a nonsense---in that mode, "apply"
279
- is not working on the data in (or derived from) the index at all.
280
- This has been fixed.
58
+ We may want to resurrect the "we don't archive an empty directory"
59
+ bonus patch, but I do not mind merging the above early to 'next'
60
+ and leave it as a separate follow-up enhancement.
61
+ cf. <20170820090629.tumvqwzkromcykjf@sigill.intra.peff.net>
62
63
--------------------------------------------------
64
[New Topics]
65
285
-* jk/drop-sha1-entry-pos (2017-08-25) 1 commit
286
- (merged to 'next' on 2017-08-25 at 7ef03bb281)
287
- + sha1-lookup: remove sha1_entry_pos() from header file
288
-
289
- Code clean-up.
66
+* dw/diff-highlight-makefile-fix (2017-09-06) 1 commit
67
+ - diff-highlight: add clean target to Makefile
68
291
- Will merge to 'master'.
69
+ Build clean-up.
70
71
+ Will merge to 'next'.
72
294
-* ls/convert-filter-progress (2017-08-24) 1 commit
295
- (merged to 'next' on 2017-08-25 at ce0bb30e8f)
296
- + convert: display progress for filtered objects that have been delayed
297
-
298
- The codepath to call external process filter for smudge/clean
299
- operation learned to show the progress meter.
300
-
301
- Will merge to 'master'.
302
-
303
-
304
-* mm/send-email-cc-cruft (2017-08-24) 2 commits
305
- (merged to 'next' on 2017-08-25 at 65933523ab)
306
- + send-email: don't use Mail::Address, even if available
307
- + send-email: fix garbage removal after address
308
-
309
- In addition to "cc: <a@dd.re.ss> # cruft", "cc: a@dd.re.ss # cruft"
310
- was taught to "git send-email" as a valid way to tell it that it
311
- needs to also send a carbon copy to <a@dd.re.ss> in the trailer
312
- section.
313
-
314
- Will merge to 'master'.
73
74
+* jk/config-lockfile-leak-fix (2017-09-06) 1 commit
75
+ - config: use a static lock_file struct
76
317
-* nd/worktree-kill-parse-ref (2017-08-24) 1 commit
318
- (merged to 'next' on 2017-08-25 at a5da82b2ea)
319
- + branch: fix branch renaming not updating HEADs correctly
77
+ A leakfix.
78
321
- "git branch -M a b" while on a branch that is completely unrelated
322
- to either branch a or branch b misbehaved when multiple worktree
323
- was in use. This has been fixed.
79
+ Will merge to 'next'.
80
325
- Will merge to 'master'.
81
82
+* kw/merge-recursive-cleanup (2017-09-06) 4 commits
83
+ - SQUASH???
84
+ - merge-recursive: change current file dir string_lists to hashmap
85
+ - merge-recursive: remove return value from get_files_dirs
86
+ - merge-recursive: fix memory leak
87
328
-* pc/submodule-helper (2017-08-25) 4 commits
329
- - submodule: port submodule subcommand 'status' from shell to C
330
- - submodule: port set_name_rev() from shell to C
331
- - submodule--helper: introduce for_each_listed_submodule()
332
- - submodule--helper: introduce get_submodule_displaypath()
88
+ A leakfix and code clean-up.
89
334
- GSoC.
90
91
+* ma/pkt-line-leakfix (2017-09-06) 1 commit
92
+ - pkt-line: re-'static'-ify buffer in packet_write_fmt_1()
93
337
-* sb/parse-options-blank-line-before-option-list (2017-08-25) 1 commit
338
- - usage_with_options: omit double new line on empty option list
94
+ A leakfix.
95
340
- "git worktree" with no option and no subcommand showed too many
341
- blank lines in its help text, which has been reduced.
96
+ Will merge to 'next'.
97
98
344
-* rs/apply-epoch (2017-08-25) 2 commits
345
- (merged to 'next' on 2017-08-26 at c2bf5ceca3)
346
- + apply: remove epoch date from regex
347
- + apply: check date of potential epoch timestamps first
99
+* ma/split-symref-update-fix (2017-09-06) 3 commits
100
+ - refs/files-backend: correct return value in lock_ref_for_update
101
+ - refs/files-backend: fix memory leak in lock_ref_for_update
102
+ - refs/files-backend: add longer-scoped copy of string to list
103
349
- Code simplification.
104
+ A leakfix.
105
351
- Will merge to 'master'.
106
+ Will merge to 'next'.
107
108
354
-* mh/notes-cleanup (2017-08-26) 12 commits
355
- (merged to 'next' on 2017-08-26 at 2d6df46966)
356
- + load_subtree(): declare some variables to be `size_t`
357
- + hex_to_bytes(): simpler replacement for `get_oid_hex_segment()`
358
- + get_oid_hex_segment(): don't pad the rest of `oid`
359
- + load_subtree(): combine some common code
360
- + get_oid_hex_segment(): return 0 on success
361
- + load_subtree(): only consider blobs to be potential notes
362
- + load_subtree(): check earlier whether an internal node is a tree entry
363
- + load_subtree(): separate logic for internal vs. terminal entries
364
- + load_subtree(): fix incorrect comment
365
- + load_subtree(): reduce the scope of some local variables
366
- + load_subtree(): remove unnecessary conditional
367
- + notes: make GET_NIBBLE macro more robust
109
+* mg/timestamp-t-fix (2017-09-06) 1 commit
110
+ - name-rev: change ULONG_MAX to TIME_MAX
111
369
- Code clean-up.
112
+ A mismerge fix.
113
371
- Will merge to 'master'.
114
+ Will merge to 'next'.
115
116
--------------------------------------------------
117
[Stalled]
@@ -448,6 +191,95 @@ of the repositories listed at
191
--------------------------------------------------
192
[Cooking]
193
194
+* jk/drop-sha1-entry-pos (2017-08-25) 1 commit
195
+ (merged to 'next' on 2017-08-25 at 7ef03bb281)
196
+ + sha1-lookup: remove sha1_entry_pos() from header file
197
+
198
+ Code clean-up.
199
+
200
+ Will merge to 'master'.
201
+
202
+
203
+* ls/convert-filter-progress (2017-08-24) 1 commit
204
+ (merged to 'next' on 2017-08-25 at ce0bb30e8f)
205
+ + convert: display progress for filtered objects that have been delayed
206
+
207
+ The codepath to call external process filter for smudge/clean
208
+ operation learned to show the progress meter.
209
+
210
+ Will merge to 'master'.
211
+
212
+
213
+* mm/send-email-cc-cruft (2017-08-24) 2 commits
214
+ (merged to 'next' on 2017-08-25 at 65933523ab)
215
+ + send-email: don't use Mail::Address, even if available
216
+ + send-email: fix garbage removal after address
217
+
218
+ In addition to "cc: <a@dd.re.ss> # cruft", "cc: a@dd.re.ss # cruft"
219
+ was taught to "git send-email" as a valid way to tell it that it
220
+ needs to also send a carbon copy to <a@dd.re.ss> in the trailer
221
+ section.
222
+
223
+ Will merge to 'master'.
224
+
225
+
226
+* nd/worktree-kill-parse-ref (2017-08-24) 1 commit
227
+ (merged to 'next' on 2017-08-25 at a5da82b2ea)
228
+ + branch: fix branch renaming not updating HEADs correctly
229
+
230
+ "git branch -M a b" while on a branch that is completely unrelated
231
+ to either branch a or branch b misbehaved when multiple worktree
232
+ was in use. This has been fixed.
233
+
234
+ Will merge to 'master'.
235
+
236
+
237
+* pc/submodule-helper (2017-08-25) 4 commits
238
+ - submodule: port submodule subcommand 'status' from shell to C
239
+ - submodule: port set_name_rev() from shell to C
240
+ - submodule--helper: introduce for_each_listed_submodule()
241
+ - submodule--helper: introduce get_submodule_displaypath()
242
+
243
+ GSoC.
244
+
245
+
246
+* sb/parse-options-blank-line-before-option-list (2017-08-25) 1 commit
247
+ - usage_with_options: omit double new line on empty option list
248
+
249
+ "git worktree" with no option and no subcommand showed too many
250
+ blank lines in its help text, which has been reduced.
251
+
252
+
253
+* rs/apply-epoch (2017-08-25) 2 commits
254
+ (merged to 'next' on 2017-08-26 at c2bf5ceca3)
255
+ + apply: remove epoch date from regex
256
+ + apply: check date of potential epoch timestamps first
257
+
258
+ Code simplification.
259
+
260
+ Will merge to 'master'.
261
+
262
+
263
+* mh/notes-cleanup (2017-08-26) 12 commits
264
+ (merged to 'next' on 2017-08-26 at 2d6df46966)
265
+ + load_subtree(): declare some variables to be `size_t`
266
+ + hex_to_bytes(): simpler replacement for `get_oid_hex_segment()`
267
+ + get_oid_hex_segment(): don't pad the rest of `oid`
268
+ + load_subtree(): combine some common code
269
+ + get_oid_hex_segment(): return 0 on success
270
+ + load_subtree(): only consider blobs to be potential notes
271
+ + load_subtree(): check earlier whether an internal node is a tree entry
272
+ + load_subtree(): separate logic for internal vs. terminal entries
273
+ + load_subtree(): fix incorrect comment
274
+ + load_subtree(): reduce the scope of some local variables
275
+ + load_subtree(): remove unnecessary conditional
276
+ + notes: make GET_NIBBLE macro more robust
277
+
278
+ Code clean-up.
279
+
280
+ Will merge to 'master'.
281
+
282
+
283
* nd/prune-in-worktree (2017-08-24) 16 commits
284
- refs.c: reindent get_submodule_ref_store()
285
- refs.c: remove fallback-to-main-store code get_submodule_ref_store()
@@ -495,23 +327,6 @@ of the repositories listed at
327
Will merge to 'master'.
328
329
498
-* rs/archive-excluded-directory (2017-08-19) 3 commits
499
- (merged to 'next' on 2017-08-22 at 1853597c35)
500
- + archive: don't queue excluded directories
501
- + archive: factor out helper functions for handling attributes
502
- + t5001: add tests for export-ignore attributes and exclude pathspecs
503
-
504
- "git archive" did not work well with pathspecs and the
505
- export-ignore attribute.
506
-
507
- Will merge to 'master'.
508
-
509
- We may want to resurrect the "we don't archive an empty directory"
510
- bonus patch, but I do not mind merging the above early to 'next'
511
- and leave it as a separate follow-up enhancement.
512
- cf. <20170820090629.tumvqwzkromcykjf@sigill.intra.peff.net>
513
-
514
-
330
* bc/hash-algo (2017-08-20) 5 commits
331
- hash-algo: switch empty tree and blob lookups to use hash abstraction
332
- hash-algo: integrate hash algorithm support with repo setup
@@ -519,36 +334,10 @@ of the repositories listed at
334
- setup: expose enumerated repo info
335
- Merge branch 'bc/vcs-svn-cleanup' into bc/hash-algo
336
522
- RFC
337
+ RFC
338
cf. <20170821000022.26729-1-sandals@crustytoothpaste.net>
339
340
526
-* ks/branch-set-upstream (2017-08-17) 3 commits
527
- (merged to 'next' on 2017-08-22 at 10fd938390)
528
- + branch: quote branch/ref names to improve readability
529
- + builtin/branch: stop supporting the "--set-upstream" option
530
- + t3200: cleanup cruft of a test
531
-
532
- "branch --set-upstream" that has been deprecated in Git 1.8 has
533
- finally been retired.
534
-
535
- Will merge to 'master'.
536
-
537
-
538
-* po/read-graft-line (2017-08-18) 4 commits
539
- (merged to 'next' on 2017-08-22 at 1e3fe0d3a1)
540
- + commit: rewrite read_graft_line
541
- + commit: allocate array using object_id size
542
- + commit: replace the raw buffer with strbuf in read_graft_line
543
- + sha1_file: fix definition of null_sha1
544
-
545
- Conversion from uchar[20] to struct object_id continues; this is to
546
- ensure that we do not assume sizeof(struct object_id) is the same
547
- as the length of SHA-1 hash (or length of longest hash we support).
548
-
549
- Will merge to 'master'.
550
-
551
-
341
* mk/diff-delta-uint-may-be-shorter-than-ulong (2017-08-10) 1 commit
342
. diff-delta: fix encoding size that would not fit in "unsigned int"
343