This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author bru
Recipients bru
Date 2014-11-25.14:30:55
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <>
The "readline" module offers "write_history_file" from readline/history.h "write_history", but there's no "append_history_file" that would invoke "append_history" from the C header.

This causes inconveniences when saving history to a file (like shown there: Indeed, say you have 2 interpreters (A) and (B) open.
- (A) and (B) load the history, composed of (z) lines
- you work on both, writing (a) lines in (A) and (b) lines in (b)
- you close (A), the history file now has (z)+(a)
- you close (B), the history file now has (z)+(b)
Therefore (A) history (the (a) lines) is lost. Offering "append_history_file" would be a nice way to fix this problem: having (z)+(a)+(b) in the end would be easy.

This is exactly what the attached patch does. With it are tests, doc and an example. I've not updated Misc/NEWS yet though (no issue #).
Date User Action Args
2014-11-25 14:30:57brusetrecipients: + bru
2014-11-25 14:30:57brusetmessageid: <>
2014-11-25 14:30:57brulinkissue22940 messages
2014-11-25 14:30:56brucreate