search

ruben2020 / codequery

A code-understanding, code-browsing or code-search tool. This is a tool to index, then query or search C, C++, Java, Python, Ruby, Go and Javascript source code. It builds upon the databases of cscope and ctags, and provides a nice GUI tool.
https://ruben2020.github.io/codequery/
Mozilla Public License 2.0
680 stars 86 forks source link

Unsupported cscope parameters: -c is required. -q, -b and -R are optional. The rest should not be used. #88

Closed d4em0n closed 7 months ago

d4em0n commented 3 years ago

Hi, i try to run cqmakedb on c project but i got the following error.

> cqmakedb -s cq.db -c cscope.out -t tags -p
cscope.out sanity check Error
Unsupported cscope parameters: -c is required. -q, -b and -R are optional. The rest should not be used.
ruben2020 commented 3 years ago

@d4em0n Did you first create a cscope.files file then run cscope like this: cscope -cb You can find this info in the HOWTO-LINUX and HOWTO-WINDOWS files.

dariomx commented 3 years ago

Greetings Ruben. Using -c option seems to considerably slow down the indexing of cscope; not sure why though. Is there any reason for not using the compressed format? (aka, without -c option).

I also wonder if the inverted index variant would be useful for codequery as well (-q option)

Thanks.

ruben2020 commented 3 years ago

@dariomx -c is used because cqmakedb does not parse the compressed version of cscope.out at the moment. But this could be a suggestion for enhancement if that really makes a difference in indexing speed.

pidgeon777 commented 2 years ago

@dariomx -c is used because cqmakedb does not parse the compressed version of cscope.out at the moment. But this could be a suggestion for enhancement if that really makes a difference in indexing speed.

It could be a nice feature to add, when handling big cscope databases.

brent-casavant-hpe commented 1 year ago

@dariomx -c is used because cqmakedb does not parse the compressed version of cscope.out at the moment. But this could be a suggestion for enhancement if that really makes a difference in indexing speed.

It could be a nice feature to add, when handling big cscope databases.

Just wanted to add an upvote for this. The project I work on the most has around 19,000 files, and anything to speed up the database build would be helpful.

ruben2020 commented 1 year ago

@brent-casavant-hpe I think I'll start looking into this again soon. It sounds like a good feature to add to cqmakedb.

ruben2020 commented 8 months ago

@brent-casavant-hpe @pidgeon777 @dariomx @d4em0n I have added the ability to support compressed cscope.out files, but it's currently experimental. It's on the develop branch. Simply leave out -c from the cscope call.

ruben2020 commented 7 months ago

@brent-casavant-hpe @pidgeon777 @dariomx @d4em0n Support for compressed cscope.out supported experimentally in v0.27.0. This issue can now be closed.