Readme update #1939
Closed
Readme update #1939
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
For example, 'SET v = {..}' will remove all properties of
v and set the provided map as its properties.
This patch is credited to deem0n (Dmitriy Dorofeev). Many thanks! Fixed a typo that causes a backup of a passed in variable to become indeterminate. This, in turn, caused the restoration of that variable to become indeterminate. Fixed a typo that caused the incorrect hash, label_name_graph_cache_hash, to be initialized instead of, label_oid_cache_hash. Added the ag_graph and ag_label tables to the pg_dump output.
* Fixed a bug where python driver throws error when build from source * Moved required packages to requirements.txt * Update Readme Rearranged instructions and added instructions to install from requirements.txt * Fixed a typo in exceptions.py file Fixed a typo 'SqlExcution' on line 62 in exceptions.py file
modified: Makefile
modified: README.md
modified: RELEASE
renamed: age--1.1.0.sql -> age--1.2.0.sql
modified: age.control
* Added comments for create_graph function
-age_exists, which is the executor for exists(property) function, was making up to 3 redundant memory accesses. -exists(property) predicate function only serves to check if a property exists or not. -In Cypher, if a property is assigned the value of NULL, this is considered the same as the property not existing at all. Thus the function calls to get the value type is not needed as the check for the null argument itself filters out all NULL inputs. If a property passes this check, it implies existence.
Additional changes: Null properties are removed from cypher maps in certain cases. A new property `cypher_map.keep_null` is added. It is set while transforming CREATE, MATCH and SET clause. Based on its value, transform_cypher_map tells executor whether to keep the null properties in maps that appear in the mentioned clauses. Cases of null removal: - For CREATE and SET with equal operator, nulls are removed. - For MATCH, nulls are not removed. - For SET with plus-equal operator, nulls are kept during transformation, but removed after execution. - For any other cases, nulls are removed by default.
- This CI workflow will run python driver unit tests and ensure that any PR/commit that modifies files under age/drivers/python is passing the tests. - Moved docker-compose.yaml file in age/drivers/golang to age/drivers,because this is required to instaniate a postgres instance, needed for unit testing. - Modified host address of connection params to localhost in 'test_age_py.py' to connect to postgres instance running in docker container.
Updated developer mailing list email
Added user mailing list subscription
Added different video tutorials in it which will help in installation of Apache Age on Windows and MacOS.
Added PostgreSQL 11 badge
- Added installcheck.yaml - Removed .travis.yaml file
* added paths directive to on: pull request * Now this CI will only run when a commit or PR has touched any file in drivers/golang
* Added comments for create_graph function
-age_exists, which is the executor for exists(property) function, was making up to 3 redundant memory accesses. -exists(property) predicate function only serves to check if a property exists or not. -In Cypher, if a property is assigned the value of NULL, this is considered the same as the property not existing at all. Thus the function calls to get the value type is not needed as the check for the null argument itself filters out all NULL inputs. If a property passes this check, it implies existence.
Additional changes: Null properties are removed from cypher maps in certain cases. A new property `cypher_map.keep_null` is added. It is set while transforming CREATE, MATCH and SET clause. Based on its value, transform_cypher_map tells executor whether to keep the null properties in maps that appear in the mentioned clauses. Cases of null removal: - For CREATE and SET with equal operator, nulls are removed. - For MATCH, nulls are not removed. - For SET with plus-equal operator, nulls are kept during transformation, but removed after execution. - For any other cases, nulls are removed by default.
- This CI workflow will run python driver unit tests and ensure that any PR/commit that modifies files under age/drivers/python is passing the tests. - Moved docker-compose.yaml file in age/drivers/golang to age/drivers,because this is required to instaniate a postgres instance, needed for unit testing. - Modified host address of connection params to localhost in 'test_age_py.py' to connect to postgres instance running in docker container.
Updated developer mailing list email
Added user mailing list subscription
Added different video tutorials in it which will help in installation of Apache Age on Windows and MacOS.
- Added installcheck.yaml - Removed .travis.yaml file
* added paths directive to on: pull request * Now this CI will only run when a commit or PR has touched any file in drivers/golang
* Updated Readme for drivers folder and added important links related to Apache-AGE project which were included in other readme files. * Updated NodeJS driver readme and added important links of Apache-AGE project as included in other readme files.
- This CI will only run whenever a PR or commit touches files drivers/jdbc. - Added logger to display the result of tests i.e. passing, failing etc.
- Removed path directive - Included PG11 as trigger branch
Found async function without any await expressions in NodeJS Driver. Async keyword is used in Asynchronous programming and for the functions that wait for some responses. These functions dont need to be Asynchronous.
- Removed path directive - Included PG11 as trigger branch
…e#1750) Added the command graph_stats. This command will force a reload of the specified graph. This is done to generate current statistics and dump any warning messages about the graph generated during a the load. Cleaned up some messaging from the VLE load process for duplicate vertices and edges. Added regression tests.
fix the typos, trivial
Implemented integer conversion to boolean in both toBoolean and toBooleanList functions. Also made corresponding changes to regression tests and broke long lines to adhere to the 79-column line length limit in the toBooleanList function.
Edited connection string to be more generic.
* Implement COUNT Subquery with PG specifications Implements a version of COUNT subquery that evaluates the underlying subquery according to Postgres subquery specifications. Modifies the grammar to accept all valid input for both COUNT and EXISTS subqueries. Add regression tests related to COUNT Modify the subquery regression tests to accomodate changes introduced in this update. * Implement Constraints on Subqueries Added logic to support constraints on subqueries. Added an expression tree walker to traverse expr trees. This logic can be expanded to search expr trees for other exprs outside of subqueries as well. Changed OPTIONAL MATCH logic to accomodate for subqueries. Changed the grammar for COUNT () functions to use a keyword. This avoids grammatical errors with other components. Add regression tests for subqueries as well as some for the expr tree walker. Reformatted previous regression tests.
…1799) Fixed issue 1767: create table as select * from cypher()... throws an error - ERROR: unhandled cypher(cstring) function call CREATE TABLE AS is a utility command and therefore is not seen by the query tree walker. This is by design, per the query_tree_walker code comments. This is why the cypher command wasn't transformed causing the above error. The fix was to add a check for utility commands in the Query section of the convert_cypher_walker. If found, the embedded query from CreateTableAsStmts nodes is extracted and processed. NOTE: This only enables the utility command CreateTableAsStmts nodes. Others will need to be added on a case by case basis. Added regression tests.
Integrated datum_image_hash and datum_image_eq to the logic for the VLE edge property datum match. Additionally, both were added to agtype_eq and agtype_ne. The hope is that, for large datums, these routines will improve performance. No impact to regression tests.
* update for psycopg3 * set a default argparse namespace in case tests are run in such a way that argparse is bypassed.
- Added a check for the case where the list comprehension is used with `WITH *`. Handled this case by adding entries in targetlist to the group clause. - Added regression tests.
- This action will now add "override-stale" label to the PRs that are by apache/age committers, this will prevent stale action from touching these PRs.
- All the PRs sent to master and PGXX branches will now require atleast 1 approving review.
- Due to the recent changes in the docker image tags, the CI workflows need to be updated to use the new tags.
Added logic to support returnless unions in subqueries. Added regression tests to support this addition. Modified regression tests to capture more cases.
- This workflow will send the first notice to the issue/PR author after 45 days of inactivity and will close the issue/PR after 7 further days of inactivity. - Only the issues/PRs after 1st Jan 2023 will be affected by this workflow. - Maximum of 500 issues/PRs can be processed in one run of this workflow. - This workflow will run after every 24 hours. - Issues/PRs with the label `override-stale` will not be affected by this workflow. - "Abondoned" label will be added to the issues/PRs which are closed by this workflow.
…che#879) Corrected typos and grammatical errors in apache-age-basic.ipynb
…pache#1881) Fixed issue 1878, an issue with delete_global_graphs during parallel regression tests. Added locking and some more defensive copies. No regression tests were impacted.
Replaces `CALL expr . expr` to `CALL expr_var . expr_func_norm`. There was a if-block before that checked if the first `expr` is a ColumnRef and the second `expr` is a FuncCall node. That check is not need now since expr_var and expr_func_norm are reduced to ColumnRef and FuncCall node respectively. This change also prevents shift\reduce ambiguity with CALL subquery syntax.
- Changed the first notice period to 60 days and closing period to 14 days. - Changed label to 'Stale' from 'Abondoned'
On Debian's s390x architecture, some regression tests were failing because the result set was reordered. Fix by attaching ORDER BY in problematic cases.
* Order some regression tests for stability on big-endian On Debian's s390x architecture, some regression tests were failing because the result set was reordered. Fix by attaching ORDER BY in problematic cases. * Move agtype_hash_cmp tests to separate file Output of the agtype_hash_cmp function varies depending on the machine endianness. Move tests to a separate file and supply little-endian and big-endian output expected files.
Fixed agtype_build_map to allow objects to be created with more than 50 pairs. This is a PostgreSQL limitation on the number of function arguments. To achieve this, the agtype_add, aka concat operator, was used to concat objects together when the limit was reached for the number of arguments. Added regression tests.
- Fix a markdown link grammar mistake. - Fix some "Age" misspellings to "AGE".
Added the function is_valid_label_name (apache#1911) per user request. This function will return a boolean value for any agtype string passed, depending on whether it is a valid name for a label. For all other types, not an agtype string, it will error out. Depending on user requirements, this could be changed in the future. Added regression tests.
Converted all single line comments in C file to multiline comments based PostgreSQL coding conventions
Fixed issue 1910: Server crashes when using exists(path) with explicit length paths. The crash was caused because transform_ColumnRef did not pass down the levels_up value into scanNSItemForColumn. Upon fixing that issue, it was found that the transform_cypher_node function did not correctly process the vle_function_end_var end variable. In some cases, in the WHERE clause, it needs to be created. Added regression tests.
…E from the $libdir/plugins directory (apache#1935) This is useful when a non-superuser login is used for the connection. Additionally, fixed a small bug where a memoryview was not transformed to bytes before decoding into a UTF-8 string. Co-authored-by: Daan Wagenaar <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I added a link to the open-source .NET driver for Apache AGE.