Fossil

Timeline
Login

Timeline

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

200 check-ins occurring around 0868adf92a81435d.

2007-12-06
03:48
Fixed handling of empty revisions. (check-in: bf0b70d5 user: aku tags: trunk)
2007-12-05
08:07
Merged mainline into my branch to get the newest application. (check-in: d0305b30 user: aku tags: trunk)
07:58
Bugfix. Translation implies encoding, not the reverse. This caused problems when parsing files with mixed-mode line-endings. The generated char offsets and lengths were off. Found during expansion. (check-in: 6f1c4424 user: aku tags: trunk)
07:57
Created new pass for the import of files and changesets. Uses the new file method and fossil accessor class to handle the file import. Changeset -> manifest conversion is _not_ covered yet. (check-in: d9fc75e5 user: aku tags: trunk)
07:55
Extended file handling with main logic to import an archive into fossil. Main parts are determining the various orders for expansion and import, and expanding all revisions in an archive into full texts, iteratively applying the stored deltas. (check-in: e100314e user: aku tags: trunk)
07:53
Integrated the new fossil management class with the option processor. New option --fossil to specify the fossil executable if it was not found in the PATH, or to overide the PATH. (check-in: 1a20a81f user: aku tags: trunk)
07:52
Added helper class managing access to fossil repositories. Already has a basic method to import and delta-compress a series of files. Used 'test' commands of fossil to get the necessary low-level access. (check-in: b6bf21e2 user: aku tags: trunk)
07:50
Put the graph traversal core of the cycle breaker core into a separate class, for use in other parts of the system. TODO: Rewrite the cycle breaker core in terms of this class. (check-in: e7013137 user: aku tags: trunk)
07:47
Extended project class with method return full path of the project, i.e. inclusive its repository base path. (check-in: 82e35d33 user: aku tags: trunk)
02:29
Went to explicit var-substitution for the dynamic sql queries, makes formatting easier. (check-in: 0ee9711e user: aku tags: trunk)
02:25
Fixed typo, although it did not break anything. (check-in: 5b2d15f1 user: aku tags: trunk)
02:24
Fixed log progress use in last breaker pass. Disabled the check regarding revision order. It still triggers and I can't see anything wrong. Pressing forward for now, i.e. get the remaining passes done, then revisit this issue. (check-in: 22905072 user: aku tags: trunk)
02:22
The handling of detached lines of development (floating branches) still had some bugs regarding the linkage to their revisions, especially the first revision on such branches. Fixed the relevant places, added early integrity checks and updated the main checks to handle the situation. (check-in: c4003e7b user: aku tags: trunk)
02:21
Removed lots of now dead code. Added a note to the last remaining user of the changeset method 'nextmap'. (check-in: 3c0ef2c3 user: aku tags: trunk)
2007-12-04
13:05
Generate CGI replies as separate header and body so that the header can be extended during the construction of the body. (check-in: 6af8fdc2 user: drh tags: trunk)
04:54
Reworked ComputeLimits in the last breaker pass. Moved the heavy computation of the max predecessor / min successor data down to the sql in the changeset class. (check-in: 711e0002 user: aku tags: trunk)
02:47
Add the timeline display preferences page with the ability to turn on and off block markup in timeline comments and to limit the length of timeline comments. (check-in: ebb27659 user: drh tags: trunk)
01:55
Fix typos in the "concepts.html" document. (check-in: ec1fbd7f user: drh tags: trunk)
01:26
Fix the revert command so that it works from subdirectories. Other minor comment and help-text changes. (check-in: 371dd657 user: drh tags: trunk)
2007-12-03
19:17
Send native deltas when cloning. The reduces bandwidth and reduces server load. (check-in: 18fbb9c5 user: drh tags: trunk)
19:08
Fix the delta computation on download so that it never tries to delta a file that has the same file as both its parent and its child. (check-in: 16fbb59b user: drh tags: trunk)
14:42
Detect delta loops and make at least one member of the loop a phantom. (check-in: fb90abe5 user: drh tags: trunk)
00:27
Patch problems with timeline. Need to rethink the whole implementation. (check-in: 20553a42 user: drh tags: trunk)
2007-12-02
23:47
Fluff: Renamed state methods use/reading/writing to usedb/use/extend for clarity. Updated all callers. Extended state module with code to dump the SQL statements it receives to a file for analysis. Extended the 'use' declarations of several passes. (check-in: e288af39 user: aku tags: trunk)
23:45
Actually put the code for the new pass into the repository. (check-in: 060513f2 user: aku tags: trunk)
22:30
Addition to the default style sheet to cover the name/value pairs at the top of the vinfo page. Also a bugfix in the report.c code. (check-in: dff17229 user: drh tags: trunk)
20:06
Importer Status...

Speed. This is now mostly acceptable. The one exception is pass 'BreakAllCsetCycles'. The reason is that the limit computation it does for backward branches still uses the inefficient file-level dependency computation. This will be tackled in short order. For the other passes the file spent is 'CsetsDeps' is recouped by the much faster graph setup.

Memory. This seems to be mostly acceptable as well, with the exceptions of 'BreakAllCsetCycles' (again, for reasons see above), and 'InitCsets'. It seems to happen while the pass breaks internal dependencies, but there is no hard data. I have to measure using a memory-debug enabled tclsh. I suspect either the actual internal dependencies, or the pseudo-dependencies. Maybe combined with a bad choice of data structures. Well, measuring first. (check-in: e8c374f6 user: aku tags: trunk)

20:04
The performance was still not satisfying, even with faster recomputing of successors. Doing it multiple times (Building the graph in each breaker and sort passes) eats time. Caching in memory blows the memory. Chosen solution: Cache this information in the database.

Created a new pass 'CsetDeps' which is run between 'InitCsets' and 'BreakRevCsetCycles' (i.e. changeset creation and first breaker pass). It computes the changeset dependencies from the file-level dependencies once and saves the result in the state, in the new table 'cssuccessor'. Now the breaker and sort passes can get the information quickly, with virtually no effort. The dependencies are recomputed incrementally when a changeset is split by one of the breaker passes, for its fragments and its predecessors.

The loop check is now trivial, and integrated into the successor computation, with the heavy lifting for the detailed analysis and reporting moved down into the type-dependent SQL queries. The relevant new method is 'loops'. Now that the loop check is incremental the pass based checks have been removed from the integrity module, and the option '--loopcheck' has been eliminated. For paranoia the graph setup and modification code got its loop check reinstated as an assert, redusing the changeset report code.

Renumbered the breaker and sort passes. A number of p... (check-in: 00bf8c19 user: aku tags: trunk)

06:58
Added progress output to the breaking of backward branches. (check-in: a437da48 user: aku tags: trunk)
06:57
Finally committing another disabled integrity constraint for changesets, explanations in the comments. (check-in: fd93aa26 user: aku tags: trunk)
06:49
Bugfix. Typo. (check-in: c7847514 user: aku tags: trunk)
06:18
Modified loop check in cycle breaker to abort immediately on first looped changeset. (check-in: 487bb1c6 user: aku tags: trunk)
06:17
Fix table linkage in query, and duplicated conditions :( (check-in: f7cca3f0 user: aku tags: trunk)
05:49
Performance bugfix. nextmap/premap can still be performance killers and memory hogs. Moved the computation of sucessor changesets down to the type-dependent code (new methods) and the SQL database, i.e. the C level. In the current setup it was possible that the DB would deliver us millions of file-level dependency pairs which the Tcl level would then reduce to tens of actual changeset dependencies. Tcl did not cope well with that amount of data. Now the reduction happens in the query itself. A concrete example was a branch in the Tcl CVS generating nearly 9 million pairs, which reduced to roughly 200 changeset dependencies. This blew the memory out of the water and the converter ground to a halt, busily swapping. Ok, causes behind us, also added another index on 'csitem(iid)' to speed the search for changesets from the revisions, tags, and branches. (check-in: 9c570550 user: aku tags: trunk)
04:55
Bugfix. Have the symbol dependency retrieval commands actually return something. (check-in: 71201058 user: aku tags: trunk)
03:49
Deactivated caching of the nextmap/premap data, with the indices the retrieval seems to be fast enough to allow us to reduce mem consumption. Tweaked log output, and sql formatting. (check-in: ac026148 user: aku tags: trunk)
03:46
Bugfix in ValidateFragments, tweaked comment a bit, bugfix in SQL, reordered tables in the successor/predecessor queries a bit to show the actual progression of their use. (check-in: fbfb5318 user: aku tags: trunk)
03:42
Fixed typo. Grr. (check-in: 0a97ad38 user: aku tags: trunk)
03:41
Added progress output to the code loading up the graph to traverse, nodes, and arcs. (check-in: 41d41c7b user: aku tags: trunk)
03:40
Added ability to declare indices on tables in the persistent state. Used this to declare indices on critical columns. Slows down the transactions saving changesets, this however is made up when it comes to successor/predecessor retrieval of changesets. (check-in: 74854a30 user: aku tags: trunk)
2007-12-01
18:36
Bugfix. Handle free-floating branches (their root revision was removed as unnecessary). (check-in: 79c227a9 user: aku tags: trunk)
18:35
Helper updated to changes in state definition. (check-in: 5bbc7d4c user: aku tags: trunk)
2007-11-30
15:21
Updated helper app to changed state definition (csrevision -> csitem). (check-in: fc635d59 user: aku tags: trunk)
15:20
Fixed typo of keyword. (check-in: 3db03be6 user: aku tags: trunk)
06:57
Easier name for self-referential changesets, loopcheck. Made conditional on option --loopcheck, default off, and avoided if the general checks on changesets report trouble. Reinstated the loop check in the cycle breaker core in simpler form, reusing the new command in the changeset class. (check-in: 0af7a3c8 user: aku tags: trunk)
06:50
Moved the integrity checks for split fragments into separate command. Reworked breaking of internal dependencies to contrain the length of the pending list. That part of the system is still a memory hog, especially for large changesets. Added notes about this and the successor retrieval being a bottleneck. (check-in: c14e8f84 user: aku tags: trunk)
04:27
Fixed bug in new changeset code, tagged and untagged item lists went out of sync. (check-in: facb4a87 user: aku tags: trunk)
03:57
Replaced the checks for self-referential changesets in the cycle breaker with a scheme in the changeset class doing checks when splitting a changeset, which is also called by the general changeset integrity code, after each pass. Extended log output at high verbosity levels. Thorough checking of the fragments a changeset is to be split into. (check-in: b42cff97 user: aku tags: trunk)
2007-11-29
09:16
Renamed state table 'csrevision' to 'csitem' to reflect the new internals of changesets. Updated all places where it is used. (check-in: 80b1e893 user: aku tags: trunk)
09:15
Fix bad variable name. (check-in: 48593049 user: aku tags: trunk)
09:14
Updated commentary regarding cycles at this point, items instead of comments, etc. (check-in: af5904e6 user: aku tags: trunk)
09:13
Extended checks for looped changesets. (check-in: 96064544 user: aku tags: trunk)
07:58
Renamed changeset method to describe modified results, and updated the one invoker (check-in: 61829b07 user: aku tags: trunk)
07:55
Renamed changeset method to describe modified results, and updated the one invoker. Modified the sorting of time ranges. Now by max, min as tiebreaker, and object name as last tiebreaker. (check-in: 04d76a9e user: aku tags: trunk)
07:50
Added missing state declarations. (check-in: 2c9c2681 user: aku tags: trunk)
07:49
Added mising state declarations. (check-in: 31d0fe7d user: aku tags: trunk)
07:47
More realignment of variable names with their content, in pass 5. (check-in: 27f093d2 user: aku tags: trunk)
07:41
Simplified some code dealing with the item -> changeset map, using the changed semantics (1:n -> 1:1). (check-in: 39e19c0c user: aku tags: trunk)
07:36
Brought the variable names into alignment with the semantics, now again naming what is stored in them. (check-in: deab4d03 user: aku tags: trunk)
07:24
Reworked the in-memory databases of changesets. Objects now hold items, not only revisions. Tags, and branches are new possibilities. Lists of ids go to the type-dependent retrieval command. List of tagged items (type/id pairs) come back, and are in the API. The 1:n map revisions to changesets is now an 1:1-map tagged items to changeset. (check-in: 0fcfbf78 user: aku tags: trunk)
07:02
Implemented time ranges and dependency retrieval for the tag and branch based changesets. (check-in: b1666f8f user: aku tags: trunk)
06:58
Moved the existing successor/predecessor code from main class to the proper singleton. Fixed config of main class, isn't simple dispatch any longer. Simplified calculation of the readable representation of changesets and removed code which has become superfluous. (check-in: 70d22835 user: aku tags: trunk)
06:47
Added equivalent checks for the tag and branch based changesets. (check-in: 4b15fa34 user: aku tags: trunk)
06:40
Shuffled and updated the existing checks of revision changesets. (check-in: de10b230 user: aku tags: trunk)
06:31
Started to adapt the integrity module. Renamed Check to CheckRev, and added analoguous commands for Tags and Branches. Added placeholders for the new integrity commands, updated main changeset integrity command. (check-in: 7c28fe13 user: aku tags: trunk)
06:23
Updated the code printing the changeset statistics to know about change to the changeset types. (check-in: 8e301242 user: aku tags: trunk)
06:21
Brought knowledge of the new types to the state definition, changed the creation of the initial changesets to use tags and branches. (check-in: 215d2f1a user: aku tags: trunk)
06:10
Integrate the new singletons with the main class, route the relevant places to them. (check-in: c74fe3de user: aku tags: trunk)
05:58
This commit starts a series of incremental changes not quite completely overhauling the handling of changesets, i.e. of project-level revisions. Actually this series of changes already started with [8ce7ffff21] as the bug it fixes was found when the first parts of the overhaul tripped the new integrity conditions for the modified changesets.

Background: In the last few days I repeatedly ran against the wall of an assertion in pass 9, last of the cycle breakers, with the revision changesets coming in out of order when the symbols were added to the dependency graph.

While walking to the office, and later re-reading the relevant parts of cvs2svn again I had several insights. Trigger was the realization that giving the tag changesets successor dependencies was wrong. Tags describe a state, they have no successors. This caused the re-read, and I recognized that my handling of the symbol changesets was completely wrong, that with using revisions as their data. It should have been the tags and branches. From there their actual dependencies (versus my reuse of revision dependencies) fell out naturally.

I have decided to commit my rewrite of the internals incrementally to make it easier to follow them, despite leaving the sourcebase in an unusable state during the series. One big commit would be much more difficult to understand.

The... (check-in: 27b15b70 user: aku tags: trunk)

00:37
Fix the merge conflict detection. (check-in: ba9af9ac user: drh tags: trunk)
2007-11-28
14:31
Remove a stray character that snuck unto the rebuild.c source file. (check-in: f088412c user: drh tags: trunk)
08:35
Bugfix in FilterSym pass. Grafting branches operated on the tags table :( (check-in: 8ce7ffff user: aku tags: trunk)
05:39
Added convenience method for assertions and used it in place of the existing if/trouble internal constructions. Changed API of 'log write' so that we can defer substituation of the message to when the write actually happen, and converted all places which would be hit by double-substitution. The remaining 'log write' calls will be converted incrementally. (check-in: 47d52d1e user: aku tags: trunk)
2007-11-27
09:07
Modified to break all backward symbols, not only branches, removed the other custom circle breaking code, should not be needed any longer (See comments for proof). (check-in: 6b520e7d user: aku tags: trunk)
09:05
Bugfix: Sort pending nodes fully deterministic, and moved to separate helper command. Tweaked log output. (check-in: 3e18606b user: aku tags: trunk)
09:04
Updated to extended changeset string, and added tabular formatting. Further tweaked output, putting timestamp adjust messages on the same line as the changeset itself. (check-in: 1c39e576 user: aku tags: trunk)
09:03
Changesets, extended human readable representation, and tweaking of log output. (check-in: 911d56a8 user: aku tags: trunk)
08:59
Bugfix in the generation of the initial symbol changesets. Keep entries apart per line-of-development. (check-in: 2e07cd71 user: aku tags: trunk)
05:08
Bugfixes when generating revision changesets. (1) The dependencies for a revision are a list, not single. (2) Use pseudo-dependencies to separate revisions of the same file from each other if they have no direct dependencies in the state. (check-in: 67876506 user: aku tags: trunk)
04:26
Continued work on the integrity checks for changesets. Moved callers out of transactions. Two checks are already tripping on bad changesets made by InitCSets (pass 5). (check-in: 8c6488de user: aku tags: trunk)
03:30
Fix a bug in the unified diff generator. (check-in: 7e9e5fea user: drh tags: trunk)
02:37
Outline for more integrity checks, focusing on the changesets. (check-in: bf83201c user: aku tags: trunk)
2007-11-26
01:33
Begin porting the CVSTrac ticket reporting code over to fossil. The new code is not yet connected into the system. The port is incomplete. (check-in: 62f37c97 user: drh tags: trunk)
2007-11-25
21:11
Add ticket configuration editing capability. (check-in: 5f3ddcc1 user: drh tags: trunk)
17:13
Changes to the diff algorithm to put bounds on run-time for very large files with many differences. (This came up on the previous check-in when you try to diff the two versions of sqlite3.c.) (check-in: 4c22ae52 user: drh tags: trunk)
16:13
Update SQLite to the latest build from CVS. Add in the FTS3 extension, though it is not yet being used. Additional work toward tickets. (check-in: f394d845 user: drh tags: trunk)
07:54
Code cleanup. Removed trailing whitespace across the board. (check-in: b679ca33 user: aku tags: trunk)
07:44
Another helper, textual, write changeset data to stdout. (check-in: 1ea319fb user: aku tags: trunk)
07:41
Tweaked log output of the topological sorter (revisions) to be tabular (aligned columns), added information (time ranges). (check-in: bcc630d3 user: aku tags: trunk)
07:39
Changesets: Added accessor for to retrieve number of changesets known, and dropped the "trunk root -> NTDB root" dependency, is problematic. (check-in: 96167b2a user: aku tags: trunk)
07:37
Fixed log output for pass 10, and added missing command. (check-in: 63052cb6 user: aku tags: trunk)
07:35
Extended cycle breaker with debug facility allowing the user to watch the neighbourhood of specific changesets during the traversal. Extended label information, highlighting of the nodes of interest. Tweaked log output a bit. (check-in: 97b4405e user: aku tags: trunk)
07:32
State log output, moved to slightly higher verbosity level to allow them to be suppressed. (check-in: 00e041dd user: aku tags: trunk)
07:31
Dot export extended, allow graph to define label color for highlighting. (check-in: eb43120a user: aku tags: trunk)
07:30
Filerevtree helper. Added missing requirements, and dialed the verbosity down. (check-in: 124fef5b user: aku tags: trunk)
03:05
Moved more parts taken over by the top. sort passes out the breaker passes, and renumbered them. (check-in: d743f04b user: aku tags: trunk)
03:00
Moved the parts taken over by the top. sort passes out the breaker passes, and renumbered them (comments). (check-in: b6b7ff79 user: aku tags: trunk)
02:59
Investigation of changeset order differences between running passes 1 to 6 and pass 6 alone show why the topological sort passes are separate in cvs2svn. The breaking of cycles can change the order of things due to different timeranges and dependencies of the broken changesets. Created two new passes for the sorting. The break passes 7 and 8 are now passes 8 and 9, and the new sort passes are 7 and 10. (check-in: 66c85b4d user: aku tags: trunk)
02:54
Tweaked log output of filtering pass a bit. (check-in: b41127b9 user: aku tags: trunk)
02:53
Tweaked human readable representation of changesets to include their type. (check-in: 0868adf9 user: aku tags: trunk)
02:52
Bugfix in pass manager, handling of open-ended pass specifications. (check-in: 9668b164 user: aku tags: trunk)
02:51
Tweaks of the log output, and reworked internals to expose not only breaking of cycles, but of paths as well. (check-in: 54e9b0a1 user: aku tags: trunk)
2007-11-24
23:59
More work on ticketing. This is a non-working incremental check-in. (check-in: a5e4e1ba user: drh tags: trunk)
21:28
Progress toward ticket editing. This is an incremental check-in. (check-in: d913179b user: drh tags: trunk)
19:33
Progress toward getting ticketing working. We can enter a new ticket and display it. Cannot yet edit a ticket. (check-in: fb358ca4 user: drh tags: trunk)
14:06
Documentation updates. (check-in: 6680679c user: drh tags: trunk)
05:31
Created convenience methods to create the human readable repesentation of a changeset and lists of such, and made liberal use of them. (check-in: 87cf6090 user: aku tags: trunk)
04:40
Bugfix in changeset class. Documented and fixed the SQL statements pulling the successor and predecessor information out of the state. It mishandled the Trunk <-> NTDB transitions. (check-in: 184c5632 user: aku tags: trunk)
04:34
Bugfix in changeset class. Forgot to update the map from revisions to containing changesets when breaking the internal dependencies of the initial changesets. This affected only the first fragment as all the revisions put into separate fragments where still pointing to the original changeset. This lead to bogus links at the level of changesets, the changeset was seemingly still referencing itself. (check-in: 17ec2d68 user: aku tags: trunk)
04:29
Bugfix in the changeset class. The index from revisions to containing changesets is not 1:1, but 1:n. While only one revision changeset is possible there can also be zero or more symbol changesets. (check-in: 8c9030e3 user: aku tags: trunk)
04:19
Tweaked log output a bit. Bugfixes: Forgot to propagate the border information into SplitRevisions, and a varname typo. (check-in: 6d636343 user: aku tags: trunk)
04:15
Added a number of assertions and must-not-happens with associated log output. Plus some small tweaks, and notes. (check-in: eabaea87 user: aku tags: trunk)
03:59
Added missing node attributes in the Replace command. (check-in: a99d5798 user: aku tags: trunk)
03:56
Updated passes 6 and 8 for to the changtes in the cycle breaker API (changed signature of save callback). (check-in: 7ed2f29d user: aku tags: trunk)
02:45
The autosync setting understands values like "on", "off", "true", and "false" in addition to 0 and 1. Updates to the documentation. (check-in: f9f7cf56 user: drh tags: trunk)
2007-11-23
23:06
Fix a bug in the default header. (check-in: 13732d49 user: drh tags: trunk)
22:09
New default header and footer and CSS. Add the ability to edit the footer. (check-in: 34af7280 user: drh tags: trunk)
12:55
Fix a redundancy in the text output of the usage comment on the tag command. (check-in: b346a7cf user: drh tags: trunk)
12:49
Remove obsolete file containing sample ticket configuration script. (check-in: 6311ce2e user: drh tags: trunk)
05:43
Modified cycle breaker API. Hook for processed nodes now takes the graph as new first argument. Extended API, exposed method for dumping the current state of the graph, including ability to dump a sub graph. (check-in: 5f5620fb user: aku tags: trunk)
05:41
Extended the dot graph exporter to allow the export of a subgraph specified through a set of nodes. Default is the export of the whole graph, as before. (check-in: 86f33190 user: aku tags: trunk)
05:38
Added two more debugging helper applications. One to extract and show a file revision tree, the other to display a dot file generated by the importer during cycle breaking passes. (check-in: 64d0019e user: aku tags: trunk)
2007-11-22
22:55
Add the %w and %W formatting options for internal printf usage. Use these formatting characters to render wiki. Fix additional problems of unterminated wiki on webpage rendering by using %w. (There are probably more problems yet to be discovered and fixed.) (check-in: 28592937 user: drh tags: trunk)
22:14
Make sure HTML tags in wiki are terminated in the diff screen too. (check-in: d6ad7aa0 user: drh tags: trunk)
22:02
Fix the rendering of comments with unclosed block HTML markup in the vinfo page. (check-in: 3afcc438 user: drh tags: trunk)
07:24
Added two helper applications which peek into a conversion state and generate graphs from it. Both helpers look at the tree of symbols and show it in toto (symbol tree), or restrict themselves to the branches, aka lines of development (lodtree). (check-in: b8c35421 user: aku tags: trunk)
07:22
Reworked the dot export internals a bit to be more general regarding labeling and attribute writing. Updated the cycle breaker to define proper labels. (check-in: f2848471 user: aku tags: trunk)
07:21
Tweaked the log output for the statistics a bit. (check-in: 2630e506 user: aku tags: trunk)
06:03
Fixed the calculation of possible parents for a detached NTDB. Has to be trunk. The symbols were skipped before, causing them to have no parents at all later on, forcing an abort after CollateSymbols. Added some high verbosity log commands for debugging of this area. (check-in: 930ec162 user: aku tags: trunk)
04:56
Fixed bug in the initialization of mybranchcode for changesets. (check-in: 47e271a4 user: aku tags: trunk)
04:21
Continued work on pass 8. Completed the handling of backward branches, file level analysis and splitting them. Extended changesets with the necessary methods to the predecessor data and proper per-revision maps. (check-in: e50f9ed5 user: aku tags: trunk)
03:47
Continued work on pass 8. Renamed 'retrograde' to 'Backward Branch', should be easier to understand, and completed the predicate testing if a branch changeset is backward or not. (check-in: 4f1b60dd user: aku tags: trunk)
03:33
Continued work on pass 8, added outline for handling of retrograde branches, extended changesets with predicate allowing us to find the branch changesets. (check-in: 4866889e user: aku tags: trunk)
03:21
Extended changeset class with in-memory database mapping from changeset ids to the proper object, and extended the objects with position information and associated accessors. Extended pass 8 to load the commit order computed in pass 6, this is stored in the new position slot of changesets, and an inverted index mapping from position to changeset at that position. (check-in: de4cff41 user: aku tags: trunk)
03:11
Modified the API for the construction of changesets a bit, now allowing their construction with the correct id, instead of correcting it later. Updated pass 5 to use this, and fixed bug where the id counter for changesets was left uninitialized, allowing the improper generation of duplicate ids. (check-in: 65be27aa user: aku tags: trunk)
03:03
Reworked the cycle breaker internals, moving the code handling the replacement of a changset (= node) with its fragments into a separate command. Extended the API, exposing the replacement operation, for use by passes. Added debugging code showing the set of consumable nodes for each iteration. (check-in: ad7d5c2d user: aku tags: trunk)
02:58
Moved the functionality for splitting a changeset based on the sets of revisions for the fragments to be into a separate command, and into the changeset class, for use outside of changeset links. (check-in: 59207428 user: aku tags: trunk)
01:53
Fix some compiler warnings. Comment changes on non-functioning code in tkt.c. (check-in: 2a707334 user: drh tags: trunk)
2007-11-21
16:03
Fix a bug in the unified diff generator. (check-in: 4db464a7 user: drh tags: trunk)
13:52
Append the pathname to the login cookie name so that separate cookies are used for each server. (check-in: d2b44699 user: drh tags: trunk)
12:21
Attach login cookies to the root path of the server, so that multiple servers can coexist on the same site. (check-in: 68a202e1 user: drh tags: trunk)
04:44
Fixed typo in cycle breaker logging. Extended pass 8, wrote the outline with all the relevant custom callbacks. (check-in: 1f600181 user: aku tags: trunk)
04:36
Cycle breaker, API change. The changesets are now communicated via a retrieval callback instead of directly. Updated passes 6 and 7. This allowed us to move the start/done graph exports into the cyclebreaker as well. Changeset selection in pass 8 now in separate command too for this. (check-in: 2a0ec504 user: aku tags: trunk)
04:11
Cycle breaker, API extension. Added a hook to process the graph between setup and regular consummation. This will be used by pass 8. (check-in: 1e177a4c user: aku tags: trunk)
04:05
Cycle breaker, API change. Routing the core cycle breaking through a callback so that users can choose their own algorithms. Updated passes 6 and 7. Moven changesets selection in pass 7 to separate command. (check-in: 2cf0462b user: aku tags: trunk)
03:46
API change cycle breaker. The save callback command is now specified through a separate configuration command. Moved callback invokation to helper command. Updated pass 6. Moved changeset selection to helper command. (check-in: d58423cd user: aku tags: trunk)
03:19
Try again to get the Login/Logout menu name correct. (check-in: f0a9e3b5 user: drh tags: trunk)
03:17
Get the name of the Login/Logout menu option right. (check-in: bd667b19 user: drh tags: trunk)
03:01
Render the header of every page using subscript. Add a setup page that for editing that subscript. (check-in: 555911df user: drh tags: trunk)
2007-11-20
23:54
Add the SbS_Render() interface for rendering HTML with embedded subscript. (check-in: e29abeff user: drh tags: trunk)
06:59
Added the ability to export the changeset graphs processed by the passes 6 to 8 using GraphViz's dot-format. This is activated by using the switch '--dots'. Bugfixes in the cycle breaker. First corrected variable names, I forgot to use the standard 'myXXX' format for the typevariables. Second, fixed a bug uncovered by looking at the exported graphs, which caused the system to loose arcs, possibly breaking cycles without actually breaking them, leaving them in the dependencies. (check-in: 7f15be90 user: aku tags: trunk)
2007-11-18
20:48
Add a more advanced commit-hook mechanism that allows us to specify multiple procedures in a particular order prior to commit. Continuing work toward getting tickets going. (check-in: d5e7891b user: drh tags: trunk)
2007-11-17
17:48
Continuing work on the infrastructure for tickets. (check-in: 8c96ed62 user: drh tags: trunk)
04:39
Added an option "--batch" with which the user can disable the feedback display standard to interactive use. (check-in: 4e7dd05f user: aku tags: trunk)
00:29
Merge aku's CVS import changes into the main line. Fix a small bug in diff.c. (check-in: 38b967dc user: drh tags: trunk)
2007-11-16
20:42
Rework the merge algorithm. It now only works for text files. But, it no longer gets confused by line endings (\r\n versus \n) and it reports conflicts. (check-in: 36b96b86 user: drh tags: trunk)
08:32
Added convenience command to the state package when the sql returns a single row. Added more statistics about revisions, tags, branches, symbols, changesets to various passes. (check-in: 96b7bfb8 user: aku tags: trunk)
06:55
Moved out-of-place state declarations to the proper phase (setup, not run). (check-in: ce7fb48e user: aku tags: trunk)
06:53
Added skeleton of pass 8, the final pass for the breaking of dependency cycles. This pass will handle the remaining cycles crossing revision and symbol changesets. (check-in: e7c805f1 user: aku tags: trunk)
06:51
Bugfix. Destroy graph object after use. (check-in: 7b392868 user: aku tags: trunk)
04:17
Completed pass 7, breaking dependency cycles over symbol changesets. Moved the bulk of the cycle breaker code into its own class as it was common to the passes 6 and 7, and updated the two passes accordingly. Added code to load the changeset counter from the state to start properly. (check-in: 770a9b57 user: aku tags: trunk)
03:59
Bugfix. When setting up or extended the changeset graph a changeset's successor may lay outside of the set of changesets under consideration, i.e. without a node in the graph. Ignore these. This did not (or only rarely) happen before the bugfix to the successor computation of changesets in project::rev (list instead of single). (check-in: de64c94f user: aku tags: trunk)
03:54
Bugfix. When I introduced the 'revisionbranchchildren' table to pass 2 I forgot to update pass 4 as well. Added code to drop the excluded revisons from this table as well, as either parent of branches, or branch child of some other revision. (check-in: 2516f4a5 user: aku tags: trunk)
03:52
Bugfix. In pass 5, loading the changesets used the type codes instead of the type names. Modified the SQL selecting the data to return the proper names. (check-in: 341d96be user: aku tags: trunk)
03:17
Bug fixes in the Myers diff algorithm. (check-in: f1b55da0 user: drh tags: trunk)
2007-11-15
21:49
Enhanced text diff subroutine uses Myers enhancements to Wagners minimum edit distance algorithm. White space at the end of lines is ignored. (check-in: 57b2735e user: drh tags: trunk)
07:17
Added skeleton files for pass 7, the breaking of dependency cycles over symbol-based changesets. (check-in: 258366a3 user: aku tags: trunk)
2007-11-14
05:26
Added note regarding 'RevisionTopologicalSortPass', which is not a separate pass for us, but part of pass 6, breaking cycles over revision changesets. (check-in: f631d438 user: aku tags: trunk)
05:11
Completed pass 6, wrote the code performing the breaking of cycles. Done by analysing each triple of changesets in the cycle at the file dependency level to see which revisions can be sorted apart. Added some additional utility routines. Extended the changeset class with the accessors required by the cycle breaker. (check-in: 94c39d63 user: aku tags: trunk)
05:08
Fixed handling of project objects when persisting them. Fill the project map. This is needed if the pass is not skipped. For the skip case we already initialize the project map when 'load'ing from the state. (check-in: 67600f77 user: aku tags: trunk)
2007-11-13
07:22
Continued work on pass 6. Completed creation of changeset graph (nodes, dependencies), started on topological iteration and breaking cycles. Basic iteration is complete, fiding a cycle ditto. Not yet done is to actually break a found cycle. Extended the changeset class with the necessary accessor methods (getting cset type, successors, time range). Note: Looking at my code it may be that my decision to save the cset order caused this pass to subsume the RevisionTopologicalSortPass of cvs2svn. Check again when I am done. Note 2: The test case (tcl repository, tcl project) had no cycles. (check-in: 85bd219d user: aku tags: trunk)
05:09
Reworked the in-memory storage of changesets in pass 5 and supporting classes, and added loading of changesets from the persistent state for when the pass is skipped. (check-in: 24c0b662 user: aku tags: trunk)
2007-11-11
00:08
Started on pass 6, breaking cycles between revision based changesets. Added skeleton files. (check-in: 2a01d504 user: aku tags: trunk)
2007-11-10
23:44
Rewrote the algorithm for breaking internal dependencies to my liking. The complex part handling multiple splits has moved from the pass code to the changeset class itself, reusing the state computed for the first split. The state is a bit more complex to allow for its incremental update after a break has been done. Factored major pieces into separate procedures to keep the highlevel code readable. Added lots of official log output to help debugging in case of trouble. (check-in: 08ebab80 user: aku tags: trunk)
20:40
Oops. pass 5 is not complete. Missed the breaking of internal dependencies, this is done in this pass already. Extended pass _2_ and file revisions with code to save the branchchildren (possible dependencies), and pass 5 and changesets with the proper algorithm. From cvs2svn, works, do not truly like it, as it throws away and recomputes a lot of state after each split of a cset. Could update and reuse the state to perform all splits in one go. Will try that next, for now we have a working form in the code base. (check-in: 95af789e user: aku tags: trunk)
07:46
Completed pass 5, computing the initial set of changesets. Defined persistent structure and filled out the long-existing placeholder class (project::rev). (check-in: 5f7acef8 user: aku tags: trunk)
05:34
Started on pass 5, computing the initial approximate set of project level revisions, aka 'ChangeSets'. Skeleton of the pass added. (check-in: 54d1e353 user: aku tags: trunk)
2007-11-09
06:42
Added note about why we omit cvs2svn's sorting passes. (check-in: 959192d4 user: aku tags: trunk)
06:28
Moved the paranoid integrity checks of pass 2 into a separate package for use by other passes, and changed pass 4 to also run these checks, albeit in a (slightly) relaxed form. See the comments at the point of call for the reason. This completes pass 4. (check-in: 131f0518 user: aku tags: trunk)
2007-11-08
16:14
Add the --detail flag to the merge command. (check-in: 50ff86af user: drh tags: trunk)
2007-11-07
22:22
Improvements to the merge algorithm so that it works better for common changes. Still more work needed. (check-in: ac6bb3ce user: drh tags: trunk)
07:46
Opcode synchronization reworked, optype table is master from which the in-memory array is loaded. (check-in: e45f47ec user: aku tags: trunk)
06:22
Pass 4 nearly completed. Rewrite symbol mutation, completed adjustment of symbol parents, added symbol refinement (noop or not flags). Now only missing piece is replication of the pass 2 paranoia checks. Note: Checks in the adjustment of parents are a bottleneck. May need indices, or something we can do in memory. (check-in: 37734390 user: aku tags: trunk)
06:19
Extended pass 2, save branch positions, needed for pass 4, and added optype table to have the revision opcodes available in the state as well. (check-in: 7ab490df user: aku tags: trunk)
2007-11-06
08:05
Continued work on pass 4. Now propagating the project level information about prefered parents down to the file level tags. Same for the file level branches is not yet done. (check-in: 3a530ef9 user: aku tags: trunk)
07:14
Continued work on pass 4. Added code bringing the file level symbol data into compliance with the project level symbol types, converting tags to branches and vice versa. (check-in: ffafc0bd user: aku tags: trunk)
06:47
Continued work on pass 4. Implemented the deletion of excluded symbols and all parts referencing them. The complex part is the regrafting of NTDB revisions should a NTDB branch be excluded. This is like 'GraftNTDB2Trunk' in 'file' when excluding everything but the trunk. (check-in: 69bf6ab9 user: aku tags: trunk)
04:57
Started on pass 4, 'FilterSymbols'. Skeleton. (check-in: 11e5d7ce user: aku tags: trunk)
04:55
Extended comment. (check-in: d314894b user: aku tags: trunk)
04:53
Fix in pass 3, do not compute prefered parents for the excluded symbols. Exclude them from the final check as well due to this. (check-in: 4c8a5a44 user: aku tags: trunk)
04:39
Completed pass 3, CollateSymbols. Added code determining for each symbol the prefered parent from all possible parents. This is the symbol with the lowest id among the set with the maximum number of occurences as a parent. (check-in: efc78b7a user: aku tags: trunk)
2007-11-05
09:04
Extended options processing to handle --exclude, --force-tag, and --force-branch options. Extended project::sym class with in-memkory databases to hold the option information and replaced the 'UserConfig' placeholder with the actual code using the new databases to determine symbol types based on user-requests. Extended the pass itself with code performing various checks on the results of type determination, partially paranoia, partially to find genuine bad requests (excluding symbols with unexcluded blockers, making a symbol with commits on it a tag, ...). NYI: Computation of the prefered parent for all symbols. (check-in: 7eaa420a user: aku tags: trunk)
02:42
Added the "e" capability for viewing ticket submitter email addresses. Additional tinkering toward the design of tickets. This check-in is only thinly tested. (check-in: 929d28e3 user: drh tags: trunk)
2007-11-03
04:39
Add support for built-in variables in subscript. (check-in: a67fbd78 user: drh tags: trunk)
04:23
Add basic math operators to subscript. (check-in: 92f6081d user: drh tags: trunk)
04:01
Subscript interpreter is now running. (check-in: 8ef26c5e user: drh tags: trunk)
2007-11-02
06:06
Continued work on pass 3, added code to determine the type of symbols based on the tag-, branch-, and commit-counts. Hook for handling data coming from the option processoris present (UserConfig), but only as a placeholder. (check-in: f888f06f user: aku tags: trunk)
04:29
Skeleton for pass 3 added, collating the symbols. (check-in: 6d4eb247 user: aku tags: trunk)
04:26
Added code for the loading of pass II data (currently only the smybols) from the state when pass II is skipped. (check-in: c3d51040 user: aku tags: trunk)
04:24
Fixed typos. (check-in: 77904c1f user: aku tags: trunk)