Skip to content

[pdf_extract]Add support for parsing pin directions#20

Open
martonmiklos wants to merge 3 commits intoRobotips:masterfrom
martonmiklos:parse_pin_io
Open

[pdf_extract]Add support for parsing pin directions#20
martonmiklos wants to merge 3 commits intoRobotips:masterfrom
martonmiklos:parse_pin_io

Conversation

@martonmiklos
Copy link
Contributor

@martonmiklos martonmiklos commented Jan 8, 2019

Many datasheets present a table with the pin number, pin name and pin
direction columns. This patch add support for the pin direction parsing.

Tested with BQ76920 datasheet:
http://www.ti.com/lit/ds/symlink/bq76920.pdf
Example:
kep
Result:
kep

Note: The patch utilize some C++11 only features. If you decide not to use these features let me know in #19 and I will update/downgrade the code.

@sebcaux
Copy link
Member

sebcaux commented Jan 8, 2019

This is a nice functionnality, but unfortunately that broke well functional extractions.
As you can see in Travis Ci or app veyor, test does not pass. You can lauch autotest directly in Qt Creator.

@martonmiklos
Copy link
Contributor Author

Auhc, I was not aware of the test thus I have not ran them. Let me fix them.

@sebcaux
Copy link
Member

sebcaux commented Jan 8, 2019

Yes I add it because with this kind of code it's easy to one thing and broke tens others !

@martonmiklos
Copy link
Contributor Author

Well, I see what is the problem: in the "package views" some pins named N/C.

I look for a third closest string for each found pin name. Having an another N/C pin name in the same line got assigned as pin direction.
kep

I will modify my loop to check all labels (including the pin number label) distance and pick the closest label.

@martonmiklos
Copy link
Contributor Author

@sebcaux please do not merge it yet. If you have a so sweet testing infrastructure in place let me add some test cases for the pin direction parsing as well.

Many datasheets present a table with the pin number, pin name and pin
direction columns. This patch add support for the pin direction parsing.
@martonmiklos
Copy link
Contributor Author

Hm it looks like we have some encoding/poppler version differences between Linux (at me all tests are passing) and the appveyor Windows builds. I will investigate it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants