[BUG]file_name argument in methods

[Tian-Jionglu]

Describe the bug
It's a bug found in my application with pycatia, to create TIFF drawing from CATDrawing.
https://github.com/Tian-Jionglu/pycatia/projects/1#card-47338604

To Reproduce

from pycatia import catia

caa = catia()
drawing_doc = caa.active_document

# dir "F:/temp/output/" is existed, dir "F:/output/" is not. "test.CATDrawing" is the active_document

# case1
drawing_doc.export_data("F:/temp/output/", "tif", overwrite=True)
    # expect to get file "F:/temp/output/test.tif". Result is com_error.

# case2
drawing_doc.export_data("F:/temp/output", "tif", overwrite=True)
    # expect to get file "F:/temp/output/test.tif". Get file "F:/temp/output.tif" as a result.

# case3
drawing_doc.export_data(f"F:/temp/output/{drawing_doc.name}", "tif", overwrite=True)
    # expect to get file "F:/temp/output/test.tif". Result is nothing

# case4
drawing_doc.export_data("F:/temp/output/test.tif", "tif", overwrite=True)
    # expect to get file "F:/temp/output/test.tif". Result is as expected.

# case5
drawing_doc.export_data("F:/temp/output/test", "tif", overwrite=True)
    # expect to get file "F:/temp/output/test.tif". Result is as expected.

# case6
drawing_doc.export_data("F:/temp/output/test.name", "tif", overwrite=True)
    # expect to get file "F:/temp/output/test.name.tif". Result is nothing.(".name" is unknown suffix, sometimes users type tihs by fault)

# case7
drawing_doc.export_data("F:/output/test", "tif", overwrite=True)
    # expect an error to remind the directory is not exist. Result is com_error

Expected behavior
see notations.

Desktop (please complete the following information):

• OS: Windows 10
• CATIA V5 version: V5-R19
• pycatia version: pycatia merged Amend document.export_data for argument 'overwrite' #60

Additional context
It's a similar bug as in #58 .
Type str or Path, the file_name argument should be?
I think the type can be uniformed in methods below:

• docuemnt.export_data
• document.save_as
• sheet.print_to_file
• maybe there is more...

[Tian-Jionglu]

A commit added to fix this issue.
More consideration is needed for question "Type str or Paht?".

[Tian-Jionglu]

One more commit added to make document.export_data work proper with relative path 55c9ab6

[Tian-Jionglu]

document.save_as do the same change via commit bbf8749

[Tian-Jionglu]

I found another solution for this issue, as the clause in documents.open:
file_name = os.path.abspath(file_name)

With relative directory supported, "Example 8" can be simpler.

#! /usr/bin/python3.6

"""
    Example 8:
    Open all CATParts in source directory and save to IGS in target directory.
"""

import os
import glob
from pycatia import CATIADocHandler

# make these directories the full pathname.
source_directory = 'tests/cat_files'
target_directory = '__junk__'

# if full paths are supplied above you should not do this.
# source_directory = os.path.join(os.getcwd(), source_directory)
# target_directory = os.path.join(os.getcwd(), target_directory)
# clauses above are no longer needed 

# This loop assumes there are NO sub-directories.
# only convert CATParts.

part_doc_pattern = os.path.join(source_directory , '*.CATPart')
files = glob.glob(part_doc_pattern)

for file in files:
    with CATIADocHandler(file) as handler:
        part_doc = handler.document
        part_doc.export_data(target_directory, 'igs', overwrite=True)

[evereux]

I've made my own changes here but have tried to incorporate what you've said. Please take a look. :-)

I don't like the use of Path.absolute() it's undocumented and I don't know if it's behavior will change or be dropped altogether for some older versions?

So, I've used the documented Path.is_absolute() to log a warning telling the user they should probably use absolute paths. I don't think pycatia should try and guess at the users intent too much.

Sorry for the delay. I've not had a chance to test what I've done here using your examples but will try and get on that this week. If com_errors are still raised we can at least provide better error handling -> messages for weird cases.

Thanks as always dude. 👍

[Tian-Jionglu]

Commit c14f338 to fix bug in former commit.

[Tian-Jionglu]

Please think about this comment.a25762e#r44547011

[evereux]

I've replied again to that comment.

Aside from that is there anything from preventing this issue being closed?