Open Hub
Black Duck Open Hub

 Settings |  Report Duplicate

0
I Use This!
Inactive

Commits : Listings

Analyzed about 5 hours ago. based on code collected about 8 hours ago.
Apr 25, 2023 — Apr 25, 2024
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
mingo query now sends a clone of itself to the iterator, so if you manipulate it, the iterator still has a copy of the query that generated it
Jay Marcyes
More... about 11 years ago
added a todo comment on lucene to change it over to ZF2 lucene instead of the 1.x version it currently uses
Jay Marcyes
More... about 11 years ago
removes cli, the code needs to be updated and I don't want to do it right now, and since I don't immediately need it, it isn't getting done, but I don't want it in the repo if it doesn't work. MingoIterator, MingoQuery are now fully fleshed out, and all the loading type methods have been removed from MingoOrm. I think this is great, it makes the MingoOrm code way simpler, and also is more straightforward to most devs, it does make adding new orms more complicated though, since the MingoQuery needs to know about MingoOrm, and vice versa, but overall, I'm happy with the change and the simplified codebase
Jay Marcyes
More... about 11 years ago
query and iterator are mainly finished and test are fleshed out for the most part. MingoOrmTest still needs some cleanup, but for the most part, all the db load stuff has been moved from MingoOrm to MingoQuery and MingoOrm now only ever will contain one row, no more of that isMulti() and stuff
Jay Marcyes
More... about 11 years ago
removes the handleCall rewrite of MingoOrm, there be dragons and I don't want to mess with it right now, I've lost 45 minutes trying to think of a better way to do it, but instead I'm just going to keep the current handleCall method, even though it is ugly
Jay Marcyes
More... about 11 years ago
checkpoint commit, halfway through pulling all iterator stuff out of MingoOrm, and having it exclusively in MingoQuery and MingoIterator
Jay Marcyes
More... about 11 years ago
adds some missing methods, and fixes some little bugs
Jay Marcyes
More... about 11 years ago
gets rid of some dead code that I didn't even remember existed. Changes the SQL base to use json instead of serialize, this is the first step in making a cross language environment (I've been toying with starting a Mingo Python project). Objects can no longer be stored in the db, this was of marginal utility to begin with, and I've only ever stored objects in one project, where we stored the Exception object, so I won't miss it.
Jay Marcyes
More... about 11 years ago
removes symfony 1.4 plugin folder since no one uses symfony 1.4 anymore
Jay Marcyes
More... about 12 years ago
adds a MingoOrm::createQuery() static method to easily get a query object with the correct values set, also updates the readme
Jay Marcyes
More... about 12 years ago
I think that has the new v3 support for MySQL, SQLite, and PostgreSQL, I need to merge it into master when I'm happy with how the Query and iterator are working
Jay Marcyes
More... about 12 years ago
adding postgres support that is like SQLite and MySQL support to get over the problems that hstore has. Turns out that Postgres is enough different than sqlite so I'm still not done with it, and sqlite will most likely have to be updated also after I finish postgres
Jay Marcyes
More... over 12 years ago
sqlite is almost done, though I'm thinking of dumping the unique id and switching _id to _rowid and just doing away with _rowid entirely, I just wanted to save my work in case I need to revert
Jay Marcyes
More... over 12 years ago
adds an object wrapper around Mingo indexes in the table, this will make it easier to add custom indexes in the future, MingoPostgres htable works but has sorting problems I just can't fix and I am knee deep in a refactor of the SQLite code, which will be followed by Postgres and MySQL refactors (those should go pretty quick though). Sadly, the new Mingo, although being way more stable and better coded, will be incompatible with existing Mingo tables (I've changed the index tables to be foreign keyed to the _rowid instead of _id because integers are tons more memory efficient than 24 char strings).
Jay Marcyes
More... over 12 years ago
Postgres hstore works, but there are certain features that just don't work (can't store arrays in the map, doesn't sort numerically on integer values. This also sets up SQLite to get upgraded with all the new code and new thoughts (I think I will make tables do foreign keys on _rowid, remove array index support and spatial index support and simplify the code where I can because there is a lot of cruft lying around).
Jay Marcyes
More... over 12 years ago
can't seem to catch a break, postgres can insert and get but limit doesn't work since arrays aren't parsed (I was using skeys(body) and svals(body) to have postgres parse the hstore, but that doesn't work with limit, so I will have to write a parser in php
Jay Marcyes
More... over 12 years ago
finally making some headway into postgres working, it can make and kill tables, and make sure the tables exist
Jay Marcyes
More... over 12 years ago
lots of changes, not alot done. Connection to the db now goes through the MingoConfig object, this frees the Interface up from having all those setUsername(), setName(), setHost() type methods. I had the idea to move the normalizeCriteria() type methods into a MingoTranslate() object, but that doesn't seem to be that clean and I'm not sure how much I like it, so I might revert back, I wish I committed yesterday, so I could just revert. Also, this branch will add Postgres support using hstore.
Jay Marcyes
More... over 12 years ago
adds MingoOrm::getInstance() method
Jay Marcyes
More... over 12 years ago
adds a MingoField::setUnique() that allows you to set that a field should be unique, this does the check by actually selecting from the db, which isn't ideal, but is better than nothing
Jay Marcyes
More... over 12 years ago
adds a test to make sure attach was working as expecting
Jay Marcyes
More... over 12 years ago
in the sql interfaces _created is now part of the main table and is indexed, this will allow you to grab _created DESC without adding a separate table. I did this because grabbing the results in reverse order is pretty common, but there was no standard way to do that.
Jay Marcyes
More... over 12 years ago
made it easier to override the default name a MingoOrm child will use in the db by adding MingoOrm::getTableName() method. This is in anticipation of moving completely over to php >= 5.3 where namespaces will make just setting the classname as the tablename impractical and so the table name will have to be set by the child class
Jay Marcyes
More... over 12 years ago
MingoOrm will now create an Orm object for each row when set() is called, and then pass that Orm representing just one row into the setOne() method, this will make it easier for manipulation of each row since you don't have to deal with the raw map
Jay Marcyes
More... over 12 years ago
changes row_id to _rowid, created to _created, and updated to _updated. This is just to standardize the magic fields that Mingo creates as it was inconsistent that there was stuff like _id and row_id. Plus, it makes the code cleaner, I much prefer get_id() and get_rowid() to getRow_id().
Jay Marcyes
More... over 12 years ago
drops the _class postfixes (this is step one of eventually moving completely to >= php 5.3 only). Moves the main codebase into the Mingo/ directory
Jay Marcyes
More... over 12 years ago
THe Orm and Interface classes should now be seralizable
Jay Marcyes
More... over 12 years ago
adds force argument to kill() so you can force an all row delete by passing in an empty MingoCriteria object. Adds test case to make sure it works. Updates CLI to allow all row deletes (Delete from table_name) to go through.
Jay Marcyes
More... over 12 years ago
fixes a bug with is* methods not correctly doing type compares (eg, an int 1 would fail against a string 1)
Jay Marcyes
More... over 12 years ago
fixes long standing bug where row_id would get trampled on update. This never got caught because we had never used row_id as the sort in an index table until just recently when a problem of row_id's getting changed to zero in the index tables and it was traced to this problem
Jay Marcyes
More... over 12 years ago