Closed changeling closed 5 years ago
Without digging in too deeply, the following changes got past these errors:
diff -r ./backend/geneaprove/importers/gedcomimport.py /Users/clarson/Desktop/geneapro-master-edited/backend/geneaprove/importers/gedcomimport.py
179c179
< last_change=date)
---
> last_change=self._process_CHAN(date))
235c235
< name=name, description=None, last_change=chan)
---
> name=name, description=None, last_change=self._process_CHAN(chan))
I haven't done a full trace of these changes, so beware.
Without digging in too deeply, the following changes got past these errors:
diff -r ./backend/geneaprove/importers/gedcomimport.py /Users/clarson/Desktop/geneapro-master-edited/backend/geneaprove/importers/gedcomimport.py 179c179 < last_change=date)
last_change=self._process_CHAN(date))
235c235 < name=name, description=None, last_change=chan)
name=name, description=None, last_change=self._process_CHAN(chan))
I haven't done a full trace of these changes, so beware.
I have fixed that issue, though a bit differently. Your patch ended up potentially calling _process_CHAN several times. I also added an automatic testsuite with your test gedcom file. It seems anonymous enough that it should not be an issue, but please let me know if you’d rather not have it there.
Emmanuel
Great! And no, please do, I anonymized that file with that thought in mind.
On Mon, Feb 11, 2019 at 2:17 AM Emmanuel Briot notifications@github.com wrote:
Without digging in too deeply, the following changes got past these errors:
diff -r ./backend/geneaprove/importers/gedcomimport.py /Users/clarson/Desktop/geneapro-master-edited/backend/geneaprove/importers/gedcomimport.py 179c179 < last_change=date)
last_change=self._process_CHAN(date)) 235c235 < name=name, description=None, last_change=chan)
name=name, description=None, last_change=self._process_CHAN(chan))
I haven't done a full trace of these changes, so beware.
I have fixed that issue, though a bit differently. Your patch ended up potentially calling _process_CHAN several times. I also added an automatic testsuite with your test gedcom file. It seems anonymous enough that it should not be an issue, but please let me know if you’d rather not have it there.
Emmanuel
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/briot/geneapro/issues/38#issuecomment-462246581, or mute the thread https://github.com/notifications/unsubscribe-auth/AAYd_jgKdYfxmIFoz-VMcBL1Qz4fH570ks5vMScngaJpZM4ax3tj .
Note that the failure occurs on the first
INDI
without aCHAN
tag:Using the following test file: test_short.ged.zip
If a
CHAN
tag is added to thatINDI
, the import then fails onsource.last_change
instead ofpersona.last_change