Show More
@@ -1,463 +1,468 b'' | |||
|
1 | 1 | # dagop.py - graph ancestry and topology algorithm for revset |
|
2 | 2 | # |
|
3 | 3 | # Copyright 2010 Matt Mackall <mpm@selenic.com> |
|
4 | 4 | # |
|
5 | 5 | # This software may be used and distributed according to the terms of the |
|
6 | 6 | # GNU General Public License version 2 or any later version. |
|
7 | 7 | |
|
8 | 8 | from __future__ import absolute_import |
|
9 | 9 | |
|
10 | 10 | import heapq |
|
11 | 11 | |
|
12 | 12 | from . import ( |
|
13 | 13 | error, |
|
14 | 14 | mdiff, |
|
15 | 15 | node, |
|
16 | 16 | patch, |
|
17 | 17 | smartset, |
|
18 | 18 | ) |
|
19 | 19 | |
|
20 | 20 | baseset = smartset.baseset |
|
21 | 21 | generatorset = smartset.generatorset |
|
22 | 22 | |
|
23 | 23 | # possible maximum depth between null and wdir() |
|
24 | 24 | _maxlogdepth = 0x80000000 |
|
25 | 25 | |
|
26 | def _walkrevtree(pfunc, revs, startdepth, stopdepth): | |
|
26 | def _walkrevtree(pfunc, revs, startdepth, stopdepth, reverse): | |
|
27 | 27 | """Walk DAG using 'pfunc' from the given 'revs' nodes |
|
28 | 28 | |
|
29 |
'pfunc(rev)' should return the parent revisions of the given 'rev' |
|
|
29 | 'pfunc(rev)' should return the parent/child revisions of the given 'rev' | |
|
30 | if 'reverse' is True/False respectively. | |
|
30 | 31 | |
|
31 | 32 | Scan ends at the stopdepth (exlusive) if specified. Revisions found |
|
32 | 33 | earlier than the startdepth are omitted. |
|
33 | 34 | """ |
|
34 | 35 | if startdepth is None: |
|
35 | 36 | startdepth = 0 |
|
36 | 37 | if stopdepth is None: |
|
37 | 38 | stopdepth = _maxlogdepth |
|
38 | 39 | if stopdepth == 0: |
|
39 | 40 | return |
|
40 | 41 | if stopdepth < 0: |
|
41 | 42 | raise error.ProgrammingError('negative stopdepth') |
|
43 | if reverse: | |
|
44 | heapsign = -1 # max heap | |
|
45 | else: | |
|
46 | heapsign = +1 # min heap | |
|
42 | 47 | |
|
43 | 48 | # load input revs lazily to heap so earlier revisions can be yielded |
|
44 | 49 | # without fully computing the input revs |
|
45 |
revs.sort(reverse |
|
|
50 | revs.sort(reverse) | |
|
46 | 51 | irevs = iter(revs) |
|
47 |
pendingheap = [] # [( |
|
|
52 | pendingheap = [] # [(heapsign * rev, depth), ...] (i.e. lower depth first) | |
|
48 | 53 | |
|
49 | 54 | inputrev = next(irevs, None) |
|
50 | 55 | if inputrev is not None: |
|
51 |
heapq.heappush(pendingheap, ( |
|
|
56 | heapq.heappush(pendingheap, (heapsign * inputrev, 0)) | |
|
52 | 57 | |
|
53 | 58 | lastrev = None |
|
54 | 59 | while pendingheap: |
|
55 | 60 | currev, curdepth = heapq.heappop(pendingheap) |
|
56 |
currev = |
|
|
61 | currev = heapsign * currev | |
|
57 | 62 | if currev == inputrev: |
|
58 | 63 | inputrev = next(irevs, None) |
|
59 | 64 | if inputrev is not None: |
|
60 |
heapq.heappush(pendingheap, ( |
|
|
65 | heapq.heappush(pendingheap, (heapsign * inputrev, 0)) | |
|
61 | 66 | # rescan parents until curdepth >= startdepth because queued entries |
|
62 | 67 | # of the same revision are iterated from the lowest depth |
|
63 | 68 | foundnew = (currev != lastrev) |
|
64 | 69 | if foundnew and curdepth >= startdepth: |
|
65 | 70 | lastrev = currev |
|
66 | 71 | yield currev |
|
67 | 72 | pdepth = curdepth + 1 |
|
68 | 73 | if foundnew and pdepth < stopdepth: |
|
69 | 74 | for prev in pfunc(currev): |
|
70 | 75 | if prev != node.nullrev: |
|
71 |
heapq.heappush(pendingheap, ( |
|
|
76 | heapq.heappush(pendingheap, (heapsign * prev, pdepth)) | |
|
72 | 77 | |
|
73 | 78 | def _genrevancestors(repo, revs, followfirst, startdepth, stopdepth): |
|
74 | 79 | if followfirst: |
|
75 | 80 | cut = 1 |
|
76 | 81 | else: |
|
77 | 82 | cut = None |
|
78 | 83 | cl = repo.changelog |
|
79 | 84 | def pfunc(rev): |
|
80 | 85 | try: |
|
81 | 86 | return cl.parentrevs(rev)[:cut] |
|
82 | 87 | except error.WdirUnsupported: |
|
83 | 88 | return (pctx.rev() for pctx in repo[rev].parents()[:cut]) |
|
84 | return _walkrevtree(pfunc, revs, startdepth, stopdepth) | |
|
89 | return _walkrevtree(pfunc, revs, startdepth, stopdepth, reverse=True) | |
|
85 | 90 | |
|
86 | 91 | def revancestors(repo, revs, followfirst, startdepth=None, stopdepth=None): |
|
87 | 92 | """Like revlog.ancestors(), but supports additional options, includes |
|
88 | 93 | the given revs themselves, and returns a smartset |
|
89 | 94 | |
|
90 | 95 | Scan ends at the stopdepth (exlusive) if specified. Revisions found |
|
91 | 96 | earlier than the startdepth are omitted. |
|
92 | 97 | """ |
|
93 | 98 | gen = _genrevancestors(repo, revs, followfirst, startdepth, stopdepth) |
|
94 | 99 | return generatorset(gen, iterasc=False) |
|
95 | 100 | |
|
96 | 101 | def _genrevdescendants(repo, revs, followfirst): |
|
97 | 102 | if followfirst: |
|
98 | 103 | cut = 1 |
|
99 | 104 | else: |
|
100 | 105 | cut = None |
|
101 | 106 | |
|
102 | 107 | cl = repo.changelog |
|
103 | 108 | first = revs.min() |
|
104 | 109 | nullrev = node.nullrev |
|
105 | 110 | if first == nullrev: |
|
106 | 111 | # Are there nodes with a null first parent and a non-null |
|
107 | 112 | # second one? Maybe. Do we care? Probably not. |
|
108 | 113 | yield first |
|
109 | 114 | for i in cl: |
|
110 | 115 | yield i |
|
111 | 116 | else: |
|
112 | 117 | seen = set(revs) |
|
113 | 118 | for i in cl.revs(first): |
|
114 | 119 | if i in seen: |
|
115 | 120 | yield i |
|
116 | 121 | continue |
|
117 | 122 | for x in cl.parentrevs(i)[:cut]: |
|
118 | 123 | if x != nullrev and x in seen: |
|
119 | 124 | seen.add(i) |
|
120 | 125 | yield i |
|
121 | 126 | break |
|
122 | 127 | |
|
123 | 128 | def revdescendants(repo, revs, followfirst): |
|
124 | 129 | """Like revlog.descendants() but supports additional options, includes |
|
125 | 130 | the given revs themselves, and returns a smartset""" |
|
126 | 131 | gen = _genrevdescendants(repo, revs, followfirst) |
|
127 | 132 | return generatorset(gen, iterasc=True) |
|
128 | 133 | |
|
129 | 134 | def _reachablerootspure(repo, minroot, roots, heads, includepath): |
|
130 | 135 | """return (heads(::<roots> and ::<heads>)) |
|
131 | 136 | |
|
132 | 137 | If includepath is True, return (<roots>::<heads>).""" |
|
133 | 138 | if not roots: |
|
134 | 139 | return [] |
|
135 | 140 | parentrevs = repo.changelog.parentrevs |
|
136 | 141 | roots = set(roots) |
|
137 | 142 | visit = list(heads) |
|
138 | 143 | reachable = set() |
|
139 | 144 | seen = {} |
|
140 | 145 | # prefetch all the things! (because python is slow) |
|
141 | 146 | reached = reachable.add |
|
142 | 147 | dovisit = visit.append |
|
143 | 148 | nextvisit = visit.pop |
|
144 | 149 | # open-code the post-order traversal due to the tiny size of |
|
145 | 150 | # sys.getrecursionlimit() |
|
146 | 151 | while visit: |
|
147 | 152 | rev = nextvisit() |
|
148 | 153 | if rev in roots: |
|
149 | 154 | reached(rev) |
|
150 | 155 | if not includepath: |
|
151 | 156 | continue |
|
152 | 157 | parents = parentrevs(rev) |
|
153 | 158 | seen[rev] = parents |
|
154 | 159 | for parent in parents: |
|
155 | 160 | if parent >= minroot and parent not in seen: |
|
156 | 161 | dovisit(parent) |
|
157 | 162 | if not reachable: |
|
158 | 163 | return baseset() |
|
159 | 164 | if not includepath: |
|
160 | 165 | return reachable |
|
161 | 166 | for rev in sorted(seen): |
|
162 | 167 | for parent in seen[rev]: |
|
163 | 168 | if parent in reachable: |
|
164 | 169 | reached(rev) |
|
165 | 170 | return reachable |
|
166 | 171 | |
|
167 | 172 | def reachableroots(repo, roots, heads, includepath=False): |
|
168 | 173 | """return (heads(::<roots> and ::<heads>)) |
|
169 | 174 | |
|
170 | 175 | If includepath is True, return (<roots>::<heads>).""" |
|
171 | 176 | if not roots: |
|
172 | 177 | return baseset() |
|
173 | 178 | minroot = roots.min() |
|
174 | 179 | roots = list(roots) |
|
175 | 180 | heads = list(heads) |
|
176 | 181 | try: |
|
177 | 182 | revs = repo.changelog.reachableroots(minroot, heads, roots, includepath) |
|
178 | 183 | except AttributeError: |
|
179 | 184 | revs = _reachablerootspure(repo, minroot, roots, heads, includepath) |
|
180 | 185 | revs = baseset(revs) |
|
181 | 186 | revs.sort() |
|
182 | 187 | return revs |
|
183 | 188 | |
|
184 | 189 | def _changesrange(fctx1, fctx2, linerange2, diffopts): |
|
185 | 190 | """Return `(diffinrange, linerange1)` where `diffinrange` is True |
|
186 | 191 | if diff from fctx2 to fctx1 has changes in linerange2 and |
|
187 | 192 | `linerange1` is the new line range for fctx1. |
|
188 | 193 | """ |
|
189 | 194 | blocks = mdiff.allblocks(fctx1.data(), fctx2.data(), diffopts) |
|
190 | 195 | filteredblocks, linerange1 = mdiff.blocksinrange(blocks, linerange2) |
|
191 | 196 | diffinrange = any(stype == '!' for _, stype in filteredblocks) |
|
192 | 197 | return diffinrange, linerange1 |
|
193 | 198 | |
|
194 | 199 | def blockancestors(fctx, fromline, toline, followfirst=False): |
|
195 | 200 | """Yield ancestors of `fctx` with respect to the block of lines within |
|
196 | 201 | `fromline`-`toline` range. |
|
197 | 202 | """ |
|
198 | 203 | diffopts = patch.diffopts(fctx._repo.ui) |
|
199 | 204 | introrev = fctx.introrev() |
|
200 | 205 | if fctx.rev() != introrev: |
|
201 | 206 | fctx = fctx.filectx(fctx.filenode(), changeid=introrev) |
|
202 | 207 | visit = {(fctx.linkrev(), fctx.filenode()): (fctx, (fromline, toline))} |
|
203 | 208 | while visit: |
|
204 | 209 | c, linerange2 = visit.pop(max(visit)) |
|
205 | 210 | pl = c.parents() |
|
206 | 211 | if followfirst: |
|
207 | 212 | pl = pl[:1] |
|
208 | 213 | if not pl: |
|
209 | 214 | # The block originates from the initial revision. |
|
210 | 215 | yield c, linerange2 |
|
211 | 216 | continue |
|
212 | 217 | inrange = False |
|
213 | 218 | for p in pl: |
|
214 | 219 | inrangep, linerange1 = _changesrange(p, c, linerange2, diffopts) |
|
215 | 220 | inrange = inrange or inrangep |
|
216 | 221 | if linerange1[0] == linerange1[1]: |
|
217 | 222 | # Parent's linerange is empty, meaning that the block got |
|
218 | 223 | # introduced in this revision; no need to go futher in this |
|
219 | 224 | # branch. |
|
220 | 225 | continue |
|
221 | 226 | # Set _descendantrev with 'c' (a known descendant) so that, when |
|
222 | 227 | # _adjustlinkrev is called for 'p', it receives this descendant |
|
223 | 228 | # (as srcrev) instead possibly topmost introrev. |
|
224 | 229 | p._descendantrev = c.rev() |
|
225 | 230 | visit[p.linkrev(), p.filenode()] = p, linerange1 |
|
226 | 231 | if inrange: |
|
227 | 232 | yield c, linerange2 |
|
228 | 233 | |
|
229 | 234 | def blockdescendants(fctx, fromline, toline): |
|
230 | 235 | """Yield descendants of `fctx` with respect to the block of lines within |
|
231 | 236 | `fromline`-`toline` range. |
|
232 | 237 | """ |
|
233 | 238 | # First possibly yield 'fctx' if it has changes in range with respect to |
|
234 | 239 | # its parents. |
|
235 | 240 | try: |
|
236 | 241 | c, linerange1 = next(blockancestors(fctx, fromline, toline)) |
|
237 | 242 | except StopIteration: |
|
238 | 243 | pass |
|
239 | 244 | else: |
|
240 | 245 | if c == fctx: |
|
241 | 246 | yield c, linerange1 |
|
242 | 247 | |
|
243 | 248 | diffopts = patch.diffopts(fctx._repo.ui) |
|
244 | 249 | fl = fctx.filelog() |
|
245 | 250 | seen = {fctx.filerev(): (fctx, (fromline, toline))} |
|
246 | 251 | for i in fl.descendants([fctx.filerev()]): |
|
247 | 252 | c = fctx.filectx(i) |
|
248 | 253 | inrange = False |
|
249 | 254 | for x in fl.parentrevs(i): |
|
250 | 255 | try: |
|
251 | 256 | p, linerange2 = seen[x] |
|
252 | 257 | except KeyError: |
|
253 | 258 | # nullrev or other branch |
|
254 | 259 | continue |
|
255 | 260 | inrangep, linerange1 = _changesrange(c, p, linerange2, diffopts) |
|
256 | 261 | inrange = inrange or inrangep |
|
257 | 262 | # If revision 'i' has been seen (it's a merge), we assume that its |
|
258 | 263 | # line range is the same independently of which parents was used |
|
259 | 264 | # to compute it. |
|
260 | 265 | assert i not in seen or seen[i][1] == linerange1, ( |
|
261 | 266 | 'computed line range for %s is not consistent between ' |
|
262 | 267 | 'ancestor branches' % c) |
|
263 | 268 | seen[i] = c, linerange1 |
|
264 | 269 | if inrange: |
|
265 | 270 | yield c, linerange1 |
|
266 | 271 | |
|
267 | 272 | def toposort(revs, parentsfunc, firstbranch=()): |
|
268 | 273 | """Yield revisions from heads to roots one (topo) branch at a time. |
|
269 | 274 | |
|
270 | 275 | This function aims to be used by a graph generator that wishes to minimize |
|
271 | 276 | the number of parallel branches and their interleaving. |
|
272 | 277 | |
|
273 | 278 | Example iteration order (numbers show the "true" order in a changelog): |
|
274 | 279 | |
|
275 | 280 | o 4 |
|
276 | 281 | | |
|
277 | 282 | o 1 |
|
278 | 283 | | |
|
279 | 284 | | o 3 |
|
280 | 285 | | | |
|
281 | 286 | | o 2 |
|
282 | 287 | |/ |
|
283 | 288 | o 0 |
|
284 | 289 | |
|
285 | 290 | Note that the ancestors of merges are understood by the current |
|
286 | 291 | algorithm to be on the same branch. This means no reordering will |
|
287 | 292 | occur behind a merge. |
|
288 | 293 | """ |
|
289 | 294 | |
|
290 | 295 | ### Quick summary of the algorithm |
|
291 | 296 | # |
|
292 | 297 | # This function is based around a "retention" principle. We keep revisions |
|
293 | 298 | # in memory until we are ready to emit a whole branch that immediately |
|
294 | 299 | # "merges" into an existing one. This reduces the number of parallel |
|
295 | 300 | # branches with interleaved revisions. |
|
296 | 301 | # |
|
297 | 302 | # During iteration revs are split into two groups: |
|
298 | 303 | # A) revision already emitted |
|
299 | 304 | # B) revision in "retention". They are stored as different subgroups. |
|
300 | 305 | # |
|
301 | 306 | # for each REV, we do the following logic: |
|
302 | 307 | # |
|
303 | 308 | # 1) if REV is a parent of (A), we will emit it. If there is a |
|
304 | 309 | # retention group ((B) above) that is blocked on REV being |
|
305 | 310 | # available, we emit all the revisions out of that retention |
|
306 | 311 | # group first. |
|
307 | 312 | # |
|
308 | 313 | # 2) else, we'll search for a subgroup in (B) awaiting for REV to be |
|
309 | 314 | # available, if such subgroup exist, we add REV to it and the subgroup is |
|
310 | 315 | # now awaiting for REV.parents() to be available. |
|
311 | 316 | # |
|
312 | 317 | # 3) finally if no such group existed in (B), we create a new subgroup. |
|
313 | 318 | # |
|
314 | 319 | # |
|
315 | 320 | # To bootstrap the algorithm, we emit the tipmost revision (which |
|
316 | 321 | # puts it in group (A) from above). |
|
317 | 322 | |
|
318 | 323 | revs.sort(reverse=True) |
|
319 | 324 | |
|
320 | 325 | # Set of parents of revision that have been emitted. They can be considered |
|
321 | 326 | # unblocked as the graph generator is already aware of them so there is no |
|
322 | 327 | # need to delay the revisions that reference them. |
|
323 | 328 | # |
|
324 | 329 | # If someone wants to prioritize a branch over the others, pre-filling this |
|
325 | 330 | # set will force all other branches to wait until this branch is ready to be |
|
326 | 331 | # emitted. |
|
327 | 332 | unblocked = set(firstbranch) |
|
328 | 333 | |
|
329 | 334 | # list of groups waiting to be displayed, each group is defined by: |
|
330 | 335 | # |
|
331 | 336 | # (revs: lists of revs waiting to be displayed, |
|
332 | 337 | # blocked: set of that cannot be displayed before those in 'revs') |
|
333 | 338 | # |
|
334 | 339 | # The second value ('blocked') correspond to parents of any revision in the |
|
335 | 340 | # group ('revs') that is not itself contained in the group. The main idea |
|
336 | 341 | # of this algorithm is to delay as much as possible the emission of any |
|
337 | 342 | # revision. This means waiting for the moment we are about to display |
|
338 | 343 | # these parents to display the revs in a group. |
|
339 | 344 | # |
|
340 | 345 | # This first implementation is smart until it encounters a merge: it will |
|
341 | 346 | # emit revs as soon as any parent is about to be emitted and can grow an |
|
342 | 347 | # arbitrary number of revs in 'blocked'. In practice this mean we properly |
|
343 | 348 | # retains new branches but gives up on any special ordering for ancestors |
|
344 | 349 | # of merges. The implementation can be improved to handle this better. |
|
345 | 350 | # |
|
346 | 351 | # The first subgroup is special. It corresponds to all the revision that |
|
347 | 352 | # were already emitted. The 'revs' lists is expected to be empty and the |
|
348 | 353 | # 'blocked' set contains the parents revisions of already emitted revision. |
|
349 | 354 | # |
|
350 | 355 | # You could pre-seed the <parents> set of groups[0] to a specific |
|
351 | 356 | # changesets to select what the first emitted branch should be. |
|
352 | 357 | groups = [([], unblocked)] |
|
353 | 358 | pendingheap = [] |
|
354 | 359 | pendingset = set() |
|
355 | 360 | |
|
356 | 361 | heapq.heapify(pendingheap) |
|
357 | 362 | heappop = heapq.heappop |
|
358 | 363 | heappush = heapq.heappush |
|
359 | 364 | for currentrev in revs: |
|
360 | 365 | # Heap works with smallest element, we want highest so we invert |
|
361 | 366 | if currentrev not in pendingset: |
|
362 | 367 | heappush(pendingheap, -currentrev) |
|
363 | 368 | pendingset.add(currentrev) |
|
364 | 369 | # iterates on pending rev until after the current rev have been |
|
365 | 370 | # processed. |
|
366 | 371 | rev = None |
|
367 | 372 | while rev != currentrev: |
|
368 | 373 | rev = -heappop(pendingheap) |
|
369 | 374 | pendingset.remove(rev) |
|
370 | 375 | |
|
371 | 376 | # Seek for a subgroup blocked, waiting for the current revision. |
|
372 | 377 | matching = [i for i, g in enumerate(groups) if rev in g[1]] |
|
373 | 378 | |
|
374 | 379 | if matching: |
|
375 | 380 | # The main idea is to gather together all sets that are blocked |
|
376 | 381 | # on the same revision. |
|
377 | 382 | # |
|
378 | 383 | # Groups are merged when a common blocking ancestor is |
|
379 | 384 | # observed. For example, given two groups: |
|
380 | 385 | # |
|
381 | 386 | # revs [5, 4] waiting for 1 |
|
382 | 387 | # revs [3, 2] waiting for 1 |
|
383 | 388 | # |
|
384 | 389 | # These two groups will be merged when we process |
|
385 | 390 | # 1. In theory, we could have merged the groups when |
|
386 | 391 | # we added 2 to the group it is now in (we could have |
|
387 | 392 | # noticed the groups were both blocked on 1 then), but |
|
388 | 393 | # the way it works now makes the algorithm simpler. |
|
389 | 394 | # |
|
390 | 395 | # We also always keep the oldest subgroup first. We can |
|
391 | 396 | # probably improve the behavior by having the longest set |
|
392 | 397 | # first. That way, graph algorithms could minimise the length |
|
393 | 398 | # of parallel lines their drawing. This is currently not done. |
|
394 | 399 | targetidx = matching.pop(0) |
|
395 | 400 | trevs, tparents = groups[targetidx] |
|
396 | 401 | for i in matching: |
|
397 | 402 | gr = groups[i] |
|
398 | 403 | trevs.extend(gr[0]) |
|
399 | 404 | tparents |= gr[1] |
|
400 | 405 | # delete all merged subgroups (except the one we kept) |
|
401 | 406 | # (starting from the last subgroup for performance and |
|
402 | 407 | # sanity reasons) |
|
403 | 408 | for i in reversed(matching): |
|
404 | 409 | del groups[i] |
|
405 | 410 | else: |
|
406 | 411 | # This is a new head. We create a new subgroup for it. |
|
407 | 412 | targetidx = len(groups) |
|
408 | 413 | groups.append(([], {rev})) |
|
409 | 414 | |
|
410 | 415 | gr = groups[targetidx] |
|
411 | 416 | |
|
412 | 417 | # We now add the current nodes to this subgroups. This is done |
|
413 | 418 | # after the subgroup merging because all elements from a subgroup |
|
414 | 419 | # that relied on this rev must precede it. |
|
415 | 420 | # |
|
416 | 421 | # we also update the <parents> set to include the parents of the |
|
417 | 422 | # new nodes. |
|
418 | 423 | if rev == currentrev: # only display stuff in rev |
|
419 | 424 | gr[0].append(rev) |
|
420 | 425 | gr[1].remove(rev) |
|
421 | 426 | parents = [p for p in parentsfunc(rev) if p > node.nullrev] |
|
422 | 427 | gr[1].update(parents) |
|
423 | 428 | for p in parents: |
|
424 | 429 | if p not in pendingset: |
|
425 | 430 | pendingset.add(p) |
|
426 | 431 | heappush(pendingheap, -p) |
|
427 | 432 | |
|
428 | 433 | # Look for a subgroup to display |
|
429 | 434 | # |
|
430 | 435 | # When unblocked is empty (if clause), we were not waiting for any |
|
431 | 436 | # revisions during the first iteration (if no priority was given) or |
|
432 | 437 | # if we emitted a whole disconnected set of the graph (reached a |
|
433 | 438 | # root). In that case we arbitrarily take the oldest known |
|
434 | 439 | # subgroup. The heuristic could probably be better. |
|
435 | 440 | # |
|
436 | 441 | # Otherwise (elif clause) if the subgroup is blocked on |
|
437 | 442 | # a revision we just emitted, we can safely emit it as |
|
438 | 443 | # well. |
|
439 | 444 | if not unblocked: |
|
440 | 445 | if len(groups) > 1: # display other subset |
|
441 | 446 | targetidx = 1 |
|
442 | 447 | gr = groups[1] |
|
443 | 448 | elif not gr[1] & unblocked: |
|
444 | 449 | gr = None |
|
445 | 450 | |
|
446 | 451 | if gr is not None: |
|
447 | 452 | # update the set of awaited revisions with the one from the |
|
448 | 453 | # subgroup |
|
449 | 454 | unblocked |= gr[1] |
|
450 | 455 | # output all revisions in the subgroup |
|
451 | 456 | for r in gr[0]: |
|
452 | 457 | yield r |
|
453 | 458 | # delete the subgroup that you just output |
|
454 | 459 | # unless it is groups[0] in which case you just empty it. |
|
455 | 460 | if targetidx: |
|
456 | 461 | del groups[targetidx] |
|
457 | 462 | else: |
|
458 | 463 | gr[0][:] = [] |
|
459 | 464 | # Check if we have some subgroup waiting for revisions we are not going to |
|
460 | 465 | # iterate over |
|
461 | 466 | for g in groups: |
|
462 | 467 | for r in g[0]: |
|
463 | 468 | yield r |
General Comments 0
You need to be logged in to leave comments.
Login now