#44 closed defect (released)
report errors on add/replace_document, not later
| Reported by: | Olly Betts | Owned by: | Olly Betts |
|---|---|---|---|
| Priority: | normal | Milestone: | |
| Component: | Library API | Version: | SVN trunk |
| Severity: | normal | Keywords: | |
| Cc: | Richard Boulton | Blocked By: | |
| Blocking: | Operating System: | All |
Description
Currently errors are sometimes not thrown on add_document/replace_document but later when an implicit or explicit flush occurs. This makes it hard to write user code to deal with such problems. Try to arrange that errors are reported sooner.
Change History (10)
comment:1 by , 20 years ago
| Severity: | blocker → normal |
|---|---|
| Status: | new → assigned |
comment:2 by , 17 years ago
| Component: | other → Library API |
|---|---|
| op_sys: | other → All |
| Priority: | high → normal |
| rep_platform: | Other → All |
| Version: | other → SVN HEAD |
comment:3 by , 17 years ago
| Blocking: | 118 added |
|---|
comment:4 by , 17 years ago
comment:5 by , 17 years ago
| Cc: | added |
|---|
+1 for postpone.
Perhaps we should have a "shortly after 1.0" tracker bug too?
comment:7 by , 17 years ago
| Blocking: | 120 added; 118 removed |
|---|
comment:10 by , 16 years ago
| Blocking: | 120 removed |
|---|
(In #120) Remove the unfixed dependencies so we can close this bug - they're all marked for the 1.1.0 milestone.
Note:
See TracTickets
for help on using tickets.
We could probably feasibly fix this for the case of termname length restrictions for terms which don't contain zero bytes for 1.0, but it doesn't change the API or ABI or database format at all so I think it's a good candidate for postponing as it could be added in 1.0.X without anyone having to relink code or rebuild databases. Thoughts?