nhoening / gritttt-rss

Implements 'cancelled' sharing-features of Google Reader for the excellent open-source RSS-Reader TinyTiny-RSS: share any page on the web via a bookmarklet and show your shared items in a widget on any website you want. Also allow to import shared&starred articles from Google Reader.
gritttt-rss.nicolashoening.de
Other
48 stars 4 forks source link

Greader import aborts #24

Closed seul closed 12 years ago

seul commented 12 years ago

Terminal output as follows ...:~/Desktop/nhoening-gritttt-rss-ca8bd9a/greader-import$ python import.py Please enter your user ID (usually, you are admin, who has ID 1): 1 Please enter the ID of your gritttt-feed (You can create a new feed with the SQL in the file 'create-gritttt-feed.sql'): 757 Should we import shared articles (Y/n)? (then I expect you to have exported a file called shared-items.json from Google): y Should we import starred articles (Y/n)? (then I expect you to have exported a file called starred-items.json from Google): y Writing gritttt-import.sql ... Reading in data from starred-items.json ... Traceback (most recent call last): File "import.py", line 131, in counter = write_sql(items, False, counter) File "import.py", line 92, in write_sql link = item['alternate'][0]['href'] KeyError: 'alternate'

gritttt-import.sql gets written, content as follows -- SQL Import from Google Reader, created 2012-06-04 21:54:54.960224

Maybe Google changed something? Files were called shared-items-jas.json and starred-items-jas.json

nhoening commented 12 years ago

I checked, and they still give you a file called starred-items.json, if you download the correct format (they have two json formats on offer). I am not sure how the README could help you more:

You need to export your shared/starred items from your Google account. Go to Google Reader -> settings, and download the shared/starred items in "Reader JSON"-format. Store the export in this directory. The files should be called shared-items.json and starred-items.json.

Maybe it can explicitly say that you should NOT download the "JSON Activity Stream" (JAS), to avoid confusion.

nhoening commented 12 years ago

Also, 757 feeds. Not bad. I have less than 200.

seul commented 12 years ago

Yes, I think »NOT the "JSON Activity Stream" (JAS)« in the readme would be a good idea. I was very tired when I started yesterday, so I just punched when I came across the first json.

Another alternative or additional possibility would be to say it in an error message – if it is at all possible for .py to easily find out if it is a JAS before doing its thing.