https://pfspipe.ipmu.jp/jira This file is an XML representation of an issue en-us 8.3.4 803005 13-09-2019 https://pfspipe.ipmu.jp/jira/browse/INSTRM-1323 Instrument control development <p><span class="error">&#91;Pulled from INSTRM-1298&#93;</span></p> <p>Please use context managers as much as possible in any database code. Both sqlalchemy and pandas support them. For sqlalchemy I believe the link is <a href="https://docs.sqlalchemy.org/en/14/orm/session_basics.html#using-a-sessionmaker" class="external-link" rel="nofollow">https://docs.sqlalchemy.org/en/14/orm/session_basics.html#using-a-sessionmaker</a> </p> <p> </p> INSTRM-1323

Use context managers ("with") as much as possible in db code.

Task Normal In Review Unresolved arnaud.lefur cloomis EngRun Tue, 13 Jul 2021 19:55:57 +0000 Thu, 18 May 2023 05:45:18 +0000 spt_operational_database 0 3 <p>It appears that significant improvements were made to context manager support between sqlalchemy 1.3.something and 1.4.something &#8211; that link does not apply to 1.3. We are running 1.3.20 at Subaru. </p> <p>The update would be clean (per <tt>mamba update sqalchemy</tt>, only sqlalchemy and greenlet would be touched).</p> <p>If people are nervous we could clone the <tt>conda-ics</tt> environment and update that for testing. I'm running one now and will report back.</p> <p>[ Note: <tt>mamba update --dry-run sqlalchemy</tt> took 15s and offered the current 1.4.21; <tt>conda update --dry-run sqlalchemy</tt> took 5 <b>minutes</b> and only offered 1.4.4. Both using the same repos. Someone's solver is wrong. ]</p> <p>I have run sqlalchemy 1.4.21 in a test conda environment for MCS testing, and it did not obviously break anything in mcs or fps. Can I update the version in <tt>conda-ics</tt>?</p> <p>I just updated the version and tested in my side and did not see any problems so far, so I think you can updete on the mountain.</p> <p>Thanks. Done (actually, 1.4.22). I believe the following idiom (from the linked document) for the Opdb class should now work:</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre># we can now construct a Session() and include begin()/commit()/rollback() # at once with self.session.begin() as session: session.add(some_object) session.add(some_other_object) # commits the transaction, closes the session </pre> </div></div> <p>I pushed a test implementation like this on the branch, but not sure if it's a right way so need a review.</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> def insert(self, tablename, dataframe):   model = getattr(models, tablename) with self.session.begin() as s: s.session.bulk_insert_mappings(model, dataframe.to_dict(orient=<span class="code-quote">"records"</span>))</pre> </div></div> <p>Can we close this ticket? If the current opDB codes (at the summit) deal with this properly, I will check them and try to make them merged.</p> <p><a href="https://pfspipe.ipmu.jp/jira/secure/ViewProfile.jspa?name=kiyoto.yabe" class="user-hover" rel="kiyoto.yabe">Kiyoto Yabe</a> I think all opDB code (as well as other software) at the summit needs to be merged to master. So we should best close this ticket after the merge.</p> <p>I think now the current master branch is (almost) compatible with the summit change. Could somebody check at summit whether it works as we expected?</p> <p>I did the most I could.<br/> I haven't tested yet, but I've mostly followed sqlalchemy prescription</p> <p><a href="https://pfspipe.ipmu.jp/jira/secure/ViewProfile.jspa?name=arnaud.lefur" class="user-hover" rel="arnaud.lefur">arnaud.lefur</a> is there any quick test you could do to close this ticket?</p> Development Rank 0|zzsx5r:c Sprint EngRun03 Story Points 2.0