J. King
4670dfc849
Handle connection errors
2019-01-15 10:51:55 -05:00
J. King
e92bda5373
Various changes:
...
- Fix handling of binary data and long strings
- Simplify handling of socket connections
- Fix coverage
2019-01-15 08:58:11 -05:00
J. King
f3b0c791f8
Fix remaining tests
2019-01-14 09:51:00 -05:00
J. King
5d61ab0a57
Fixes for MySQL native interface
...
Three test failures remain, but these are minor and will be resolved
soon. Handling of binary data is also broken, but given that this works
fine with the PDO driver, there is presumably some correct method.
2019-01-13 23:17:19 -05:00
J. King
e501fbdc87
Remove the DatabaseInformation class in tests and use traits instead
2019-01-12 12:43:06 -05:00
J. King
81acba90dc
Use strict equality when comparing strings
2019-01-11 10:38:06 -05:00
J. King
c4a41255b0
Experimental native MySQL driver
...
No testing has been performed yet, but changes are extensive enough to
warrant a commit. Of particular note:
- SQL states are enumerated in a separate trait to reduce duplication
- PDOStatement is now an abstract class to avoid duplication of
engine-specific error handling
- Error handling has been cleaned up somewhat
2019-01-10 19:01:32 -05:00
J. King
206cca35a9
Test tweaks
2018-12-21 17:51:49 -05:00
J. King
f0d30c2eee
Make munging of queries a generic feature
2018-12-21 12:35:10 -05:00
J. King
24df564045
Mostly successful MySQL database function tests
...
Two failures remain, at least one requiring query munging.
2018-12-21 10:14:26 -05:00
J. King
4ef36643a4
Proof-of-concept PDO MySQL driver
...
- Configuration options were added
- Non-transactional locking was added to the savepoint handlers
- Tests were adjusted for MySQL's reserved words
2018-12-20 18:06:28 -05:00
J. King
29e7c1f154
Fix coverage
2018-12-13 19:56:07 -05:00
J. King
2bebdd44cf
Implementation of native PostgreSQL interface
...
Changes to the Database class were required to avoid outputting booleans
2018-12-13 19:47:51 -05:00
J. King
b52dadf345
Make existing PostgreSQL tests explicitly PDO tests
2018-12-12 12:42:40 -05:00
J. King
28f803dd28
Handle PostgreSQL connection errors
2018-12-12 11:15:07 -05:00
J. King
73729a6be8
Simplify database cleanup between tests
2018-12-10 13:17:04 -05:00
J. King
8dbf237626
Group PostgreSQL tests as slow
2018-12-10 12:39:09 -05:00
J. King
913cf71620
Fix incorrect annotations
2018-12-07 20:36:20 -05:00
J. King
f6966659a9
Use smarter coverage executer; properly suppress stderr during CLI tests
2018-12-07 20:25:48 -05:00
J. King
0513b606c2
Merge master
2018-12-07 19:21:44 -05:00
J. King
089f666de6
Fix PDO insert ID errors in PHP 7.1
2018-12-06 17:46:00 -05:00
J. King
cf896121b2
Style fixes
2018-12-05 17:28:11 -05:00
J. King
f2245861e3
Restore complete Database coverage
...
Also suppress PostgreSQL database function tests from normal coverage,
and add a "coverage:full" task to run them if needed.
2018-12-05 17:07:47 -05:00
J. King
51755a2ce6
Retire article field groups
2018-12-05 16:55:14 -05:00
J. King
0129965bbd
Cover some missed code
2018-12-05 12:54:19 -05:00
J. King
22941f5ad1
Fix session tests
...
PostgreSQL now passes all tests. Connection and permission errors still
need to be accounted
for before the implementation is complete.
2018-12-05 12:07:45 -05:00
J. King
258be1d54e
Fix most PostgreSQL test failures
...
Reasons for failures included an unhandled error code, erroneous sorting
assumptions, and a broken computation of the next insert ID in tests
Five failures remain.
2018-12-05 09:05:43 -05:00
J. King
8fc31cfc40
Rewrite various queries to work in PostgreSQL
...
This involved changes to the driver interface as well as the database
schemata. The most significantly altered queries were for article
selection and marking, which relied upon unusual features of SQLite.
Overall query efficiency should not be adversely affected (it may have
even imprved) in the common case, while very rare cases (not presently
triggered by any REST handlers) require more queries.
One notable benefit of these changes is that functions which query
articles can now have complete control over which columns are returned.
This has not, however, been implemented yet: symbolic column groups are
still used for now.
Note that PostgreSQL still fails many tests, but the test suite runs to
completion. Note also that one line of the Database class is not
covered; later changes will eventually make it easier to cover the line
in question.
2018-12-04 20:41:21 -05:00
J. King
527ecee393
Code coverage fixes
2018-11-29 13:56:15 -05:00
J. King
4a1c23ba45
Munge PostgreSQL queries instead of adding explicit casts
...
PDO does not adequately inform PostgreSQL of a parameter's type, so type
casts are required. Rather than adding these to each query manually, the
queries are instead processed to add type hints automatically.
Unfortunately the queries are processed rather naively; question-mark
characters in string constants, identifiers, regex patterns, or geometry
operators will break things spectacularly.
2018-11-29 13:45:37 -05:00
J. King
e68fcc0afa
Manipulate only those sequences in the current PostgreSQL schema
2018-11-28 17:16:03 -05:00
J. King
4a2efd9987
Correct the state of PostgreSQL serial sequence during tests
2018-11-28 16:24:12 -05:00
J. King
dd4f22e04e
Avoid use of reserved SQL word "user"
2018-11-28 14:21:36 -05:00
J. King
10b228224d
Correct PostgreSQL data format and other tweaks
2018-11-28 12:12:49 -05:00
J. King
8dfedd30ef
Test PostgreSQL schema upgrade
...
This was in fact buggy due to the schema version check causing an error
2018-11-28 10:46:23 -05:00
J. King
93af381436
Test setting of schema name
2018-11-27 17:39:39 -05:00
J. King
1414f8979c
Fix savepoint handling and locking in PostgreSQL driver
2018-11-27 17:16:00 -05:00
J. King
8a49202036
Use common cleanup code for all database-related tests
2018-11-27 14:26:33 -05:00
J. King
925560d4ba
Cleanup
2018-11-25 00:06:20 -05:00
J. King
a75fad53ca
Adapt the rest of the test series
2018-11-25 00:03:56 -05:00
J. King
dccd4caede
Convert one database function test series (articles) to a common harness
...
Also revert the dropping of tables in the schema files. This was for the
convenience of tests, but the risk of data loss is too great
2018-11-24 23:18:17 -05:00
J. King
7340d65c0e
Make data clearing in tests static
2018-11-23 10:01:17 -05:00
J. King
39110858b7
Move database function test series as first step in re-organization
2018-11-23 09:29:06 -05:00
J. King
8c20411359
Align statement tests with other database driver tests
2018-11-22 23:18:20 -05:00
J. King
f22e53fdc9
Align result tests with driver tests
2018-11-22 19:55:54 -05:00
J. King
aa1b65b5d4
Take a different tack on shared database tests
...
Tests for different drivers will have their own files, but all derive
from a common prototype test series where applicable, similar to the
existing arrangement for database function tests. However, the prototype
will reside with other test cases rather than in the library path. The
database function test series will hopefully be moved as well in time.
2018-11-22 13:55:57 -05:00
J. King
736a8c9d0c
Improved timeout handling for both SQlite and PostgreSQL
2018-11-22 13:30:13 -05:00
J. King
4e444fd86c
Generic database interface creation in tests
2018-11-21 13:06:01 -05:00
J. King
c0c4810662
Nominally complete PostgreSQL driver
...
Connection error handling as well as uprade error handling still need
to be implemented.
2018-11-21 11:06:12 -05:00
J. King
84b4cb7465
Enable PostgreSQL statement testing
...
Tests involving binary data are skipped for now
2018-11-20 16:32:18 -05:00
J. King
b5733b070c
Clean up statement tests
...
PostgreSQL tests are suppressed for now, but most pass.
2018-11-20 15:45:20 -05:00
J. King
976672de5b
Test cleanup
2018-11-16 21:32:27 -05:00
J. King
edfae438fa
Refine pg connection strings
2018-11-16 21:20:54 -05:00
J. King
c4ca9149a1
Move Statement tests
2018-11-09 15:01:46 -05:00
J. King
7e11019e83
Consolidate statement tests
2018-11-09 14:58:10 -05:00
J. King
ea6b4c951e
Skip Result tests when necessary
2018-11-09 14:56:30 -05:00
J. King
a3dbb08da9
Remove obsolete test code
2018-11-08 16:29:46 -05:00
J. King
30d6f6db37
Consolidate Db result test series into single file
2018-11-08 14:50:58 -05:00
J. King
3b3b810f10
Replicate some unspecified NCNv1 behaviour, and document the rest
...
See #139 for list
Closes #139
Closes #140
Closes #141
2018-11-07 13:01:46 -05:00
J. King
63ae6fb703
Merge remote-tracking branch 'remotes/origin/user-rewrite'
2018-11-06 16:36:50 -05:00
J. King
3a4100576a
Merge remote-tracking branch 'remotes/origin/cli-overhaul'
2018-11-06 16:36:35 -05:00
J. King
39134f5f7e
Rest of CLI tests
2018-11-06 16:35:33 -05:00
J. King
4869559fb3
Test NCNv1 user query
2018-11-06 13:21:53 -05:00
J. King
2dd1b45d3e
Cover the Query class with database tests for now
2018-11-06 12:51:34 -05:00
J. King
ba8e208d79
Partial CLI tests
2018-11-06 12:32:28 -05:00
J. King
a8cc9a4780
Tests for internal user driver; closes #50
2018-11-04 12:06:30 -05:00
J. King
d40243a84c
Clean up configuration setting in tests
2018-11-04 09:16:34 -05:00
J. King
1ac85df46b
Last set of tests for User class
2018-11-03 13:26:22 -04:00
J. King
b8f8a617fe
Simply user test data providers
...
The user manager no longer differentiates between the internal driver
and other drivers, making the duplication unnecessary
2018-11-02 17:28:12 -04:00
J. King
a52b985826
Cover the Query class with database tests for now
2018-11-02 12:14:46 -04:00
J. King
ffa7bd5a5d
Fix error in previous commit
2018-11-02 12:01:03 -04:00
J. King
931fe3b585
Move password generation to the User class
...
This allows user drivers which wish to generate their own passwords to
do so, and those which do not to defer to the built-in generator
2018-11-02 11:52:55 -04:00
J. King
5959c0672d
Tests for most of the User class
2018-11-02 10:02:37 -04:00
J. King
11747c93fd
Strip out unused user management functionality
...
Tests have been removed as well; new tests are forthcoming
2018-10-28 10:59:17 -04:00
J. King
0be9dcb5a8
Update dependencies
2018-10-26 16:27:18 -04:00
J. King
e08d82f855
Style fix
2018-10-26 14:58:36 -04:00
J. King
f64f0c6a22
Whitespace cleanup
2018-10-26 14:58:04 -04:00
J. King
1aa556cf12
Add HTTP authentication support to TTRSS; fixes #133
...
Also bump version to 0.4.0
2018-10-26 14:40:20 -04:00
J. King
b4b2b10db3
Cover fatal exception; improves #66
2018-08-17 10:34:54 -04:00
J. King
edbfb12d17
Improve coverage slightly
...
Improves #66
2018-08-17 08:35:13 -04:00
J. King
cbc9491f75
CS fixes and version bump
2018-01-12 09:48:33 -05:00
J. King
34b508171b
Merge CORS branch
2018-01-12 09:41:13 -05:00
J. King
1a4989e1e0
Merge PDO branch
2018-01-12 09:28:58 -05:00
J. King
cef061f6cd
Fix tests
2018-01-11 16:00:56 -05:00
J. King
aa57227097
Use PSR-7 for authentication; fixes #53
2018-01-11 15:48:29 -05:00
J. King
daea0ceb27
Make HTTP message testing generic
...
assertMessage will test the method and target URL of requests, attributes of server requests, as well as the existing testing of a response's status code. All messages' bodies and header fields are tested for equivalence (with a special case for JSON response bodies).
2018-01-11 11:09:25 -05:00
J. King
90dfeb727a
Implement CORS; fixes #126
2018-01-09 12:31:40 -05:00
J. King
0ec0a5b085
Ensure the request method is always uppercased
2018-01-08 17:11:38 -05:00
J. King
4b53c5e8b3
Tests and fixes for REST class; fixes #53 ; improves #66
2018-01-07 12:59:10 -05:00
J. King
3fa2d38f31
Initial rewrite of REST class; needs more testing, but should be functional
...
- improves #53
- improves #66
2018-01-06 12:02:45 -05:00
J. King
890f9b07d4
Replace Resquest objects with PSR-7 request messages; improves #53
2018-01-04 23:08:53 -05:00
J. King
9ad0b47201
Add Target class to manipulate request traget URL parts
...
The query part is not parsed for now because PSR-7 request objects/PHP take care of that parsing for us.
2018-01-04 14:06:45 -05:00
J. King
9eadd602bd
Replace Response objects with PSR-7 response messages; improves #53
...
While the test suite passes, this commit yields a broken server: replacing ad hoc request objectss with PSR-7 ones is still required, as is emission of PSR-7 responses. Both will come in subsequent commits, with tests
Diactoros was chosen specifically because it includes facilities for emitting responses, something which is awkward to test. The end of this refactoring should see both the Response and Request classes disappear, and the general REST class fully covered (as well as any speculative additions to AbstractHanlder).
2018-01-03 23:13:08 -05:00
J. King
27caf147df
Changes to Date helper class
...
- Changed 'transform' method to use ValueInfo throughout. This fixes a number of obscure bugs
- Changed the 'add' and 'sub' methods to default to "now" rather than null. This means null passes through rather than being interpreted as the current time, to be consistent with other date tools
- Also changed the 'add' and 'sub' methods so that they operate correctly with invalid date strings
- Added tests for the class; improves #66
- Modified TTRSS tests because the "iso8601" format string in ValueInfo is different from Date's older format
2018-01-02 16:53:38 -05:00
J. King
89bfc23d32
Standardize date normalization to immutables
...
Also move date formats to the ValueInfo class
Standardizing on immutables avoids any possible ambiguity in the API of the resultant value, as well as any ambiguity as to whether a DateTime output instance is the same instance or a clone (they had been clones)
2018-01-02 16:27:58 -05:00
J. King
ddf55e0665
Change session lifetimes to more closely match TTRSS
...
At least some clients seem to expect the default timeout of one day.
2018-01-01 12:31:42 -05:00
J. King
65b08d1735
Tweak
2017-12-31 17:30:30 -05:00
J. King
77793f95cb
Controller fixes for PDO databases; fixes #72
2017-12-31 17:24:40 -05:00
J. King
029c23d0cf
Adapt API tests to allow for PDO mocks
2017-12-31 13:58:37 -05:00
J. King
ba0aeab7ec
Make SQL statement type conversion use ValueInfo normalizer
...
This sees the addition of a dateOutFormat parameter to ValueInfo::normalize(), as well as a general simplification of how parameter binding works.
Some value type-casting results are slightly different, but this simply makes SQL statement objects consistent with the rest of the system.
2017-12-30 18:50:56 -05:00
J. King
bc9fcb975f
Change "rebind" to "retype" to better reflect what actually happens
2017-12-30 17:04:21 -05:00
J. King
31aea0a06a
PDO-SQLite type binding tests
...
Db namespace is now back to full coverage
2017-12-30 16:16:59 -05:00
J. King
bc9ffa0e17
Rewrite SQLite 3 type binding tests
2017-12-30 15:59:45 -05:00
J. King
39cad91b78
CS fixes
2017-12-22 11:51:58 -05:00
J. King
a5318d1b12
Part 2
2017-12-22 11:45:24 -05:00
J. King
095fe10aec
Reorganize PDO tests into namespaces
2017-12-22 11:41:54 -05:00
J. King
59fee52074
Reorganize tests into namespaces
2017-12-21 22:47:19 -05:00
J. King
0785f832de
CS fixes
2017-12-19 22:19:42 -05:00
J. King
4bada691e9
PDO tests and fixes; improves #72
2017-12-19 19:08:08 -05:00
J. King
ad6a09ffa1
Partially tested pdo_sqlite driver; improves #72
2017-12-19 17:15:05 -05:00
J. King
ef75b5e9ab
SQLite driver tweaks
2017-12-19 12:11:49 -05:00
J. King
73c8583744
Remove reliance on xdebug; fixes #113
...
Code coverage information is now gathered via phpdbg (a separate executable) by if available rather than xdebug, as the latter is hard to turn on and off.
A "test:quick" task has also been added to Robo, which excludes 31 tests which together account for almost two thirds of the test run time. This should pave the way for testing to be added as a commit hook for Git.
2017-12-17 10:27:34 -05:00
J. King
183718204d
Grouped slow tests; improves #113
2017-12-16 20:03:04 -05:00
J. King
dd7d6e44b9
Further fixes for test timezone assumptions
2017-12-16 19:37:30 -05:00
J. King
3f65625090
Ensure tests do not rely in UTC default timezone
...
Six tests failed if the default timezone was not UTC. These were all due to faulty expectations, however, not faulty output.
2017-12-16 18:30:45 -05:00
J. King
415f7fd385
Fix feed timeout test; improves #113
2017-12-16 17:21:23 -05:00
J. King
3e42fbdddf
Munge off-by-one dates in tests; fixes #112
2017-12-08 16:00:23 -05:00
J. King
11f4eed118
Scripts for Robo
2017-12-08 15:15:20 -05:00
J. King
4bc3398157
Use Robo for programming task execution
...
The plan had originally been to use Bldr, but Bldr's features were not particularly compelling, and it has not been maintained recently. By contrast Robo has very useful features while still meeting my needs.
In addition to migrating the build process, the current Robo file also runs unit tests, with or without coverage reporting.
Fixes #116 ; improves #113
2017-12-08 14:37:49 -05:00
J. King
3a07156259
Fix broken test
2017-12-07 19:39:32 -05:00
J. King
0b3b3cb49c
Use SQLite nocase collation; improves #127
2017-12-07 18:05:34 -05:00
J. King
9f4095f00d
Test server tweak
2017-12-07 17:27:42 -05:00
J. King
3adfe8742f
Reorganize test cases
2017-12-07 15:46:49 -05:00
J. King
e6607453bc
Disentangle Composer dev dependencies
2017-12-07 15:09:03 -05:00
J. King
84bd624e94
Properly fix getCounters
2017-12-06 19:16:35 -05:00
J. King
adece521a9
TTRSS: Correct mark toggling; fixes #132
2017-12-01 22:13:27 -05:00
J. King
9ac2421fe3
TTRSS: accept base64 passwords; fixes #130
2017-11-30 19:31:40 -05:00
J. King
3ffcd6dd97
Make TTRSS handler reject erroneous paths
2017-11-30 17:54:56 -05:00
J. King
a404d4d108
NCNv1: Be explicit about types; fixes #129
...
This is actually mandated by the specification: a change in data type requires a protocol version number change
2017-11-30 14:47:39 -05:00
J. King
4d4e8e3580
TTRSS: correct feed renaming; fixes #128
2017-11-30 12:49:23 -05:00
J. King
22cdc8916d
Documentation tweaks and CS fixes
2017-11-29 22:42:50 -05:00
J. King
5cd7268c0a
Added incorrectDbCharset to NCNv1 server status
...
This has been exposed since version 11.0.3, released before our version 0.1.0, but after implementation work had begun
2017-11-29 18:14:59 -05:00
J. King
40944a9b58
Merge remote-tracking branch 'remotes/origin/ttrss'
2017-11-29 16:05:23 -05:00
J. King
3d958547a5
Implement OPTIONS requests for NCNv1; fixes #107
2017-11-29 15:28:33 -05:00
J. King
e1f1c8b859
Refactoring
2017-11-29 13:41:26 -05:00
J. King
2bbc83aeb0
Tweak
2017-11-29 12:15:37 -05:00
J. King
40e9b7f986
Changed all TTRSS outputs to match original types exactly; improves #125
2017-11-29 11:47:10 -05:00
J. King
b4890eaced
Make TTRSS tests consider data types; improves #125
2017-11-29 09:22:59 -05:00
J. King
91165cdd0d
Fix breakage caused by last comit
2017-11-27 15:05:50 -05:00
J. King
b820a004d6
Complete testing of TTRSS handler
...
Also implemented OPTIONS handling for TTRSS; improves #107
2017-11-23 18:07:56 -05:00
J. King
a61aa0a22c
Simplify TTRSS test request boilerplate
2017-11-23 14:05:26 -05:00
J. King
c669273792
Implement TTRSS operation getHeadlines; fixe #82
2017-11-22 20:18:16 -05:00
J. King
faf00d63ba
Add Database::articleCategoriesGet()
...
This method retrieves author-supplied categories for articles, used in TTRSS
2017-11-21 09:22:58 -05:00
J. King
5c140aedc4
Implement TTRSS operation getCompactHeadlines; fixes #95
...
This commit also implements the back-end for the standard getHeadlines operation and handles all special feeds and categories; fixes #119
2017-11-20 00:09:20 -05:00
J. King
5d4ea6edc0
Add ResultEmpty class
...
This allows for the creation of synthetic empty result sets
2017-11-19 15:49:41 -05:00
J. King
2037efce61
Added oldestArticle and latestArticle context options
2017-11-18 16:06:49 -05:00
J. King
50185ab8f6
Change code and type of consantUnknown exception
2017-11-17 22:53:54 -05:00
J. King
b0da9a1d06
Added annotation context to satisfy the TTRSS "has_note" view mode
2017-11-17 19:08:35 -05:00
J. King
b595815eb9
Change NCNv1 handler to use new "typical" articleList fieldset
2017-11-17 18:12:00 -05:00
J. King
c88b5c4f18
Make distinction between modified and marked for articles
...
Also added various sets of fields to include when listing articles.
2017-11-17 17:52:00 -05:00
