feat: convert from deprecated setup.py to pyproject.toml #15
Conversation
this pull-request will: 1. Introduce a pyproject.toml needed due to deprecation of setup.py 2025-10-31 2. Keep a small setup.py file to handle compilation of c libraries
|
@blex-max new pull-request with using pyproject.toml |
|
@blex-max what do you think about this new setup? |
|
Hello, apologies, we're really busy at the moment, but I appreciate that you want to get moving on this. Which systems have you tested this approach on? I can confirm this builds and compiles on my arm64 mac, but I haven't tested usage at all. Though in principle the changes look fairly trivial and safe |
|
N.B. If we merge this I think in principle this constitutes a major version bump as old install methods will no longer work |
I have only tested it on my system where it builds: $ python3 --version
Python 3.12.11 | packaged by conda-forge | (main, Jun 4 2025, 14:45:31) [GCC 13.3.0] on linux
$ pip --version
pip 25.0.1
$ arch
x86_64
$ uname -a
Linux E1-056347.science.psu.edu 6.14.0-33-generic #33~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Fri Sep 19 17:02:30 UTC 2 x86_64 x86_64 x86_64 GNU/Linux
|
|
Are you able to test that this build runs as you expect on data? I expect that it will, but since there is no CI or tests available in this repo it's likely worth checking |
|
I have tested parabam with data now, using the parabam subset --input ~/Downloads/test.bam --rule get_mapped_reads.py I also tried to run telomerecat, where parabam is used. That also works telomerecat bam2length test.bam
telomerecat bam2telbam test.bamWhat I haven't tested is |
Smeds
force-pushed
the
convert-setup-to-pyproject.toml
branch
from
c3554d5 to
58562da
Compare
|
@blex-max I added a gitaction workflow to test Don't have a test for the |
…ADERS]] [-v {0,1,2}]
[--temp_dir DIR] --rule RULE --input INPUT [INPUT ...] [--debug]
[--pair] [--coord] [-r [REGION]] [-d] [--output [OUTPUT]]
optional arguments:
-h, --help show this help message and exit
-p [P] The maximum amount of processes you wish
parabam stat to use. This should be less
than or equal to the amount of processor
cores in your machine [Default: 4].
-s [S] The amount of reads considered by each
distributed task. [Default: 250000]
-f [READERS] The amount of open connections to the file being read.
Conventional hard drives perform best with
the default of 1. [Default: 1]
-v {0,1,2} The amount of information output by the program:
0: No output [Default]
1: Total Reads Processed
2: Detailed output
--temp_dir DIR Path for parabam stat to use for intermediate files [Default: /tmp].
--rule RULE, -i RULE The file containing the rule, written in python,
that we wish to apply to the input BAM.
--input INPUT [INPUT ...], -b INPUT [INPUT ...]
The file(s) we wish to operate on.
Multiple entries should be separated by a single space
--debug Only the first 5million reads will be processed
--pair A pair processor is used instead of a conventional processor
--coord Engines recieve a list of reads which all map to the same starting position.
This mode is not compatible with the `--pair` option
-r [REGION], --region [REGION]
The process will be run only on reads from
this region. Regions should be colon separated as
specified by samtools (eg 'chr1:1000,5000')
-d parabam will not process reads marked duplicate.
--output [OUTPUT], -o [OUTPUT]
Specify a name for the output CSV file. If this argument is
not supplied, the output will take the following form:
parabam_stat_[UNIX_TIME].csv
[Status] parabam is quitting gracefully
|
@blex-max I have added tests for |
|
@blex-max Ran the |
|
@blex-max should we bump up the version of the software in this pull-request or in a separate one? |
|
@blex-max would it be possible to take one more look at this pull-request soon? It now has gitactions to test run the tool |
|
@blex-max bump |
|
It would be super nice if we could have this pull-request handled soon! |
|
@blex-max when the pull-request is handled it would be awesome to soon have a new release! |
|
Hello, I can see this is important to you but the reality is that |
|
Good to know! Does that mean that https://github.com/cancerit/telomerecat is also unsupported since it depends on parabam. My plan was to incorporate telomerecat this into the galaxyproject, I got a request from a user to do it (that currently is using telomerecat). My urgency is more related to that I will be ending my position here at the end of the year and after that I will probably not be doing much more work on this issue. |
|
That is a good question that I can't immediately answer, but I will raise internally. I'm not aware that telomerecat is unsupported, certainly that was never indicated on the repo (unlike parabam), but it hasn't been looked at in a while. Parabam being unsupported doesn't necessarily imply that telomerecat is unsupported, but clearly we need to have a think about that. @AndyMenzies |
2 participants
this pull-request will:
Should solve issues: #14, #10