I use Mercurial in my project and we just switched to "force text" in the Asset Serialization option.
My files are text now, but when I try to merge, I get this "tool opendiff can't handle binary" message. I tried to force FileMerge to merge and it marks the whole file as different.
I tried to switch to DiffMerge and force it to merge, and it tells an important info: "File has a NUL byte at offset 0; this is not valid for encoding..."
The file bash command tells me that those asset files are "text/plain; charset=us-ascii".
hg diff between the two revisions results in a YAML text file, so both files are in text mode.
Isn't that a bug? I mean, Unity should convert to text files that merge tools are able to merge...
By the way, we are using Mac OSX.
Don't try to convert the project in more than one repository and after merge them.
Somehow it worked when I converted the project on one repo and distributed it to others, that is, no problem in future merges.
The downside is that everybody has to get to the same revision, stop to work and wait for the conversion.
Related
we are working on a project, where the angularjs web project is compiled and binaries are stored in hg repo. The problem is angularjs js files are usually compiled with hashing for all binary files. Ex: binary files are suffixed with unique extensions for each file
main.1cc794c25c00388d81bb.js,
polyfills.eda7b2736c9951cdce19.js,
runtime.a2aefc53e5f0bce023ee.js,
common.7d1522841bf85b01f2e6.js,
1.620807da7415abaeeb47.js,
2.93e8bd3b179a0199a6a3.....etc.
The problem is every time a new binary in checkin in hg repo, it is being detected as new file and retained along with old file of same name. So, i need a way to fool the hg repo, to retain the file name but still consider them as old file replacing the previous one.
main.1cc794c25c00388d81bb.js ==> overwrite old main.js
polyfills.eda7b2736c9951cdce19.js ==> overwrite old polyfill.js
runtime.a2aefc53e5f0bce023ee.js ==> overwrite old polyfill.js
common.7d1522841bf85b01f2e6.js ==> overwrite old commom.js
1.620807da7415abaeeb47.js ==> overwrite old 1.js
2.93e8bd3b179a0199a6a3 ==> overwrite old 2.js
Could any one point out a way, to fool the hg to consider these files are just modification of previous files and not as new files ?
Can hgignore or some other extension be used...
A VCS shall track the state of files. And those are indeed new files. One can argue that those are the old files renamed - which can be recorded by the VCS.
So there are two solutions I see:
Record moving the old filenames to the new filenames. hg addremove --similarity XX might be of big help here. It will result in all the files having the new names each time - but if the similarity is good enough it will work nicely. You might need to adjust the XX to get a similarity measure (0 ... 100) which works for you best. Adding --dry-run for testing purposes might make testing easy. You WILL need to delete the old files before you run hg addremove though.
Have a pre-commit hook which iterates over *.js files and moves via an appropriate regex ..js to *.js omitting the hashing code, effectively overwriting the generic filenames with the newly generated hashed filenames.
In my mercurial repository I have a lot of different files. For example .org and .tex files. Exporting the .org file generates a .tex file. Normally I do not want to ignore .tex files since these are usually source files. But I am looking for something like:
foo.org present => ignore foo.tex
foo.org absent => do not ignore foo.tex
There is no such option available in Mercurial.
What you could do is create an extension of your own to make some kind of 'hg customstatus' command available. This could be a wrapper around the 'hg status' command that executes the filtering you want.
Git has the cat-file command to inspect internal files, e.g. git cat-file blob 557db03 will show the contents of the object whose hash starts with 557db03.
Are there similar tools for mercurial that allow me to look at all the different data files that merfcurial uses internally?
Try hg --debug help and you can see the list of all the debug commands:
debugancestor:
find the ancestor revision of two revisions in a given index
debugbuilddag:
builds a repo with a given DAG from scratch in the current empty repo
debugbundle:
lists the contents of a bundle
debugcheckstate:
validate the correctness of the current dirstate
debugcommands:
list all available commands and options
debugcomplete:
returns the completion list associated with the given command
debugdag:
format the changelog or an index DAG as a concise textual description
debugdata:
dump the contents of a data file revision
debugdate:
parse and display a date
debugdiscovery:
runs the changeset discovery protocol in isolation
debugfileset:
parse and apply a fileset specification
debugfsinfo:
show information detected about current filesystem
debuggetbundle:
retrieves a bundle from a repo
debugignore:
display the combined ignore pattern
debugindex:
dump the contents of an index file
debugindexdot:
dump an index DAG as a graphviz dot file
debuginstall:
test Mercurial installation
debugknown:
test whether node ids are known to a repo
debugpushkey:
access the pushkey key/value protocol
debugrebuildstate:
rebuild the dirstate as it would look like for the given revision
debugrename:
dump rename information
debugrevlog:
show data and statistics about a revlog
debugrevspec:
parse and apply a revision specification
debugsetparents:
manually set the parents of the current working directory
debugstate:
show the contents of the current dirstate
debugsub:
(no help text available)
debugwalk:
show how files match on given patterns
debugwireargs:
(no help text available)
There are a lot of them, and they pretty much expose everything.
The closest commands would be:
hg cat -r rev aFile
hg cat: Print the specified files as they were at the given revision
This is not completely the same than git cat-file though, as the latter can also list SHA1, type, and size for a list of objects.
In that second case, hg manifest might be more appropriate.
I want Mercurial to ignore any file with a certain extension.
For example, I wanted to ignore files with a .SUO extension. (There's no need to version-control Visual Studio user settings.)
So I changed my .hgignore file to this:
syntax: glob
*.suo
However, this has no effect, and Mercurial still sees my .suo file.
What am I doing wrong here?
If, when running hg status before altering your .hgignore file, the .suo file had a ? in front of it, then it should be ignored now. If anything else (M or A for example) it is already tracked by the repository and will not magically stop being tracked. In such a case you'll need to do hg remove on the file to delete it and have hg stop tracking it, or just do hg forget on it to have hg stop tracking it but keep the file. Either should be followed by a commit.
The only files that will be omitted from the status listing if their path matches a pattern in the .hgignore file are files that are not tracked. It would make no sense to omit a file that is tracked, because you would never see whether it had been modified, added, or removed.
Edit: Mercurial does only track files (you can't make it track empty directories), but the patterns in .hgignore are simply run against strings of the file paths relative to the root of the repository. The very same relative paths that it shows you when you run hg status. So it does work how you say you want it to work because the following lines are a standard part of my own .hgignore files:
syntax: glob
*\obj\*
*\bin\*
*.csproj.user
*.suo
Again, when you run hg status and it shows a .suo file, what single character is at the beginning of that line? Is it a M, A, R, ! or ? character? What is the path after it?
Mercurial uses entries in a file called .hgignore to determine what files it completely ignores. It is normally located in the root file for your repository (and not in the .hg directory, which you might think).
You can find out more here:
http://www.selenic.com/mercurial/hgignore.5.html
Normally, we use regular expression syntax to ensure that case is not a factor in extensions:
# use regexp syntax.
syntax: regexp
(?i)\.dcu
(?i)\.identcache
(?i)\.dof
(?i)\.dsk
(?i)\.bak
(?i)\.old
That way, it ensures that even if for some reason the case of the extension changes, it is still ignored.
Example for ignoring/excluding files with .o extension:
.*\.o$
should translate to .*\.suo$ for .suo extensions.
I have used this method successfully
Check where .hgignore file is located and ensure it is either in $HOME or project root folder. Check the CASE (vs case) of the extension. I doubt if pattern matching is case insensitive.
edit: tested, the pattern matching is NOT case sensitive. Hence, add "*.SUO" if you want to ignore files with ".SUO" extension.
My question is very similar to this one.
I have the files under Mercurial version control and each file have a string:
<modified-time>some time</modified-time>
This line have created the an external tool, and the tool change this line periodically. I don't want to know about changes in this line and want to ignore the changes in this line when I doing 'diff', 'commit' and 'status' commands.
So, if a file has only one change and the change is that line I need to skip this file in the output of 'hg status' command.
If the file has another changes in the other lines I need to get this file in the output of 'hg status' command.
If I commit the file I need to commit all the changes.
(UPDATE sorry to misread the question and mislead OP.)
If all your files are generated by the external tool, track the source files instead.
If not, and if you can modify the generator, make it use the date keyword as with the bundled keyword extension.
If not (OP's comment indicated so), you could use pre-* hooks to ignore the time change (when it's the only change). It'd be a pain, though, since you have to have a hook for every command that would see the file difference.
Or as OP's comment suggested, revert the file when it has only time change, either manually or automatically (on a timer or subscribe to file system notification).