There are a number of cases where queries in both pgadmin4 and pgadmin3 are done against pg_depend but falis to constrain the query on classid.
In particular, if for example a constraint and a sequence exists with the same oid (which is perfectly valid, as they are in different tables), a column will suddenly refer sequences that are completely incorrect. When we look up sequences, we have to make sure we only match dependencies against pg_class.
case. Its purpose is now to handle certain scinarios specific to table
children modules.
Renamed it to 'TableChildSwitchCell' from 'SwitchDepCell' to avoid any
conflict in future.
1) In SQL pane, the signature in comments had arguments with name and type, Now it will show arguments with type only.
2) There is no indent+newline between arguments (there isn't even a space). - Fixed
3) The default value for the last argument is omitted. - Fixed
Last argument is omitted because the value for the last argument we got from table is empty.
Now we form function signature in the get_definition.sql using 'pg_catalog.pg_get_function_arguments' method.
Above changes will reflect in Functions, Procedures & Trigger function Nodes.
- On droping a synonyms - the query should be modified in SQL pane.
- A proper message should be given on trying to drop an already dropped
synonyms.
- Handle the synonyms with special characters.
Fixes#1607, #1608, #1609
algorithm.
This patch takes care of:
* Consistent behaviour during create, update operation on any node.
- It should return the node data during creating a new object, or
updating the existing one.
* Now that - we have consistent behaviour during these operations, we
can consistently move, update the tree item based the node data from
the server.
* Implemented the methods for each node to get the information about
that particular node only.
* Using the above changes to implement the 'refresh' operation on tree
node properly.
I must thank Surinder, and Harshal for helping me implement the 'node'
method, and also changed the behaviour of create & update methods to
return the node data for most of the nodes.
Fixes#1244
Made backend changes for:
* Taking care of the connection status in the psycopg2 driver. And, when
the connection is lost, it throws a exception with 503 http status
message, and connection lost information in it.
* Allowing the flask application to propagate the exceptions even in the
release mode.
* Utilising the existing password (while reconnection, if not
disconnected explicitly).
* Introduced a new ajax response message 'service_unavailable' (http
status code: 503), which suggests temporary service unavailable.
Client (front-end) changes:
* To handle the connection lost of a database server for different
operations by generating proper events, and handle them properly.
Removed the connection status check code from different nodes, so that
- it generates the proper exception, when accessing the non-alive
connection.
Fixes#1387
1. The user will specify the tablespace path in test_config.json.in
2. If tablespace path not found, skip the test cases for that server(Only tablespace test cases)
3. Add the skipped test summary in the test result. (Now it's showing on console + in log file, but need to update in a final enhanced test summary report. Which is research point we will work on that after finishing all nodes API test cases)
4. Removed the test_ prefix from the values in the config files.
5. Add tablespace and roles tests
1] datamodel.js: For collection type added check before adding model into "changed" list instead of adding it blindly.
2] Type casting from str to int of column properties like attlen, attpricision.
3] Added missing data formating when adding new column to existing table (from table edit mode).
4] Added more validation for Foreign key constraint.
5] Column.js: Column grid (in table edit mode) show proper disabled color for disabled cell when grid renders.
6] All constraints and index js added idattribute to distinguish which s/constraint or s/index is updated on server side.
7] Column update.sql: Fixed sql when altering column data type. The issue was when we alter data type from which has length and precision to data type which don't have these properties and vice versa.
For e.g. alter data type numeric(10,12) to real (and vice versa)
8] Renaming constraint (RM1500).
9] simplejson KeyError handing for Python 2.7.
- Test framework support API testing with multiple server for this we need to modify test_config.json(for user it’s test_config.json.in) and test_advanced_config.json(for user it’s test_advanced_config.json.in). Server details of PG and PPAS are included in both .in files.
- Removed the logic of logging in the test client on each test scenario(As per Khushboo's comment in previous email). We need this logic in test cases under ‘browser/tests/’ as for test scenarios like change password and invalid login test cases as test client should be logged out first. So, as per this the code is slightly modified in ‘browser/tests/’.
Without quotes, the property can be mistakenly treated as a regex:
e.g.: URL: /pgadmin/browser/,
When that happens, JS engine throws "SyntaxError: invalid regular expression flag b" and the whole web interface freezes.
This problem usually occurs when pgadmin is hosted inside a subdirectory (https://host/pgadmin/ instead of https://host/).
Root cause:-
When user change schema of already created fdw extension then while creating the new fdw we should append the <schema_name>.<fdw_handler_name>.
Previously we added only <fdw_handler_name> so while executing the query it will not be able to find fdw handler name without schema name.
e.g. If user changed the fdw handler 'postgres_fdw_handler' under schema 'test' then we should display the handler name under fdw as "test. postgres_fdw_handler".
Solution:-
With this patch, we have added <schema_name>.<fdw_handler_name> so while creating or updating the fdw, proper sql will be generated.
1. RM#1293 - SQL pane is not displaying GRANT queries in functions node
2. RM#1294 - Comments are not visible in sql pane for trigger functions and functions
Explanation:
Previously we we using 'get_defintion.sql' template to generate SQL for SQL pane for functions, procedure & trigger functions node.
but GRANT statements and COMMENTS changes were missing. In order to fix that, now we have used both 'create.sql' and 'get_definition.sql' templates to generate full SQL query for SQL pane.
1. Altered variable control to make its UI consistent with privileges and Security labels.
2. Changed datamodel.js to handle duplicate rows at datamodel level and not UI/Control level. (See variable control for example)
1. Select2 control fixed for multi-select mode.
2. Index constrains (primary and unique key) fixed column listing issue.
3. Table node "inherited from" (Select2 control) was not showing table name with public schema
1. Foreign keys listed twice
The label for Exclusion constraint was set to 'Foreign keys'. fixed it.
2. Trigger function, Unique constraint, Primary key should be in title case
3. if user click outside of Preferences window it disappear (it should not disappear)
set preference alertify dialog option closableByDimmer to false.
4. Columns node settings listed twice
Override the register_preferences method in Columns CatalogObjectColumnsModule class
because there are two types of columns node, catalogs columns and tables columns.
so the settings for table columns should also work for catalog columns. Thats why we are not showing preference option for catalogs column node.
Ashesh fixed this issue.
When not using qtLiteral, it was result into an error - "unknown
error displayed when modify comments which may special characters on
Materialized view".
* Do not need to translate an empty string. (that results into the
translation header inclusion in the javacript module).
* String.prototype.StartsWith is not an well received function, it has
not been available in the QWebkit for linux.
The following node module script will be loaded, when first database
node object is added in the browser tree.
* Domains
* Foreign tables
* FTS Configuration
* FTS Dictionary, and
* Functions
Icons for these modules needs to be shown in the dependents list.
procedure node.
This patch also includes other fixes related 'procedure' node as below:
* Generate correct statement for dropping procedure.
* Fetch the missing schema name for the EXEC script.
* Removed the 'DROP CASCADE' menu options, as it is not supported by
procedure.
* Hide the security tab for the objects under the catalog schema.
Patched by Surinder Kumar, vastly improved by me.
* Make the securtiy label across all the nodes, using the same backbone
model across the nodes, and make changes in the backend code for the
same.
* Added 'catalog' as parent type for the domain, procedure nodes.
* Proper initialization of the FTS Configuration, FTS Parser, FTS
Dictionary, FTS Template Nodes.
* Proper Indentation in collection.js
* Moved the Columns, and Constraint to its repective tabs.
* Ensure all the labels only have a capital letter on the first word
(except for keywords or acronyms).
* Resolve the typo in string - 'System table'.
* Error messages on fields should not be shown unless the field loses
focus and has an error.
* "General" section should have a limited subset of information.
* Variables grids should not be on the Security tab.
* Fixes the field labels that imply a question.
* Privileges controls on the Properties lists should be in a "Security"
group.
Apart from review comments improved few other areas too:
* Used in-place DepsCell functionality for columns in both index, and
type node.
* Improved the error handling in constraint nodes.
* Added create, alert SQL Help for nodes associated with the table node.
- Table, Column, Index and Trigger (Author:- Murtuza Zabuawala)
- Constraints Primary Key/Unique/Check/Foreign Key/Exclusion
with integration into Table node (Author:- Harshal Dhumal)
- Rule (Author:- Surinder Kumar)
- Vacuum Control (Initial patch by Surinder Kumar and further enhancement by Murtuza)
In order to resolve the issue, set the height, width of the area (div)
under the wcFrame. wcIFrame consider the height, and width of this
container to determine the height, width & position of the iframe.
tree nodes.
Each individual node is responsible for introducing the 'canDrop'
function/flag to decide whether the 'Delete/Drop' context menu should be
enabled/disabled.
In commit-id: 26aa5607ad, 'obj.canDrop'
was set to true in the 'pgBrowser.Node' in delete callback, just to make
the server-group droppable, which is wrong, as all the nodes are
getting affected because of this change in a wrong way.
To fix the issue, added the 'canDrop' function in the server-group node.
Apart from them, also added restriction for not allowing to delete the
default server-group.
Also, handled the same restriction at the server end.
Tweaked by Ashesh Vashi to integrate the backgroud process, and also
with some improvements as stated below:
* Resolved an issue loading existing preference.
* Improved the background process observer/executor for supporting
detalied view.
* Added the utility path preferences in the ServerType class.
'postgis_tiger_geocoder'.
Current implementation changes the schema name, whenever we chance the
name, which is wrong. We should have not change the schema for this
event.
Thanks Fahar Abbas for reporting.
operation like backup, restore, etc within it.
Also:
* improvised the color combination of the background process logger.
* Removed an unnecessary print statement from the
get_storage_directory(..) function, also return None if STORAGE_DIR
is set to None.
1) Grant wizard close issue.
2) Moved grant wizard specific css from wizard.css to grant_wizard.css
3) removed 'ajs_content' css from wizard.css causing padding issue in alertify dialog
We will be using the external utilities like pg_dump, pg_dumpall,
pg_restore in background. pgAdmin 4 can be run as a CGI script, hence -
it is not good idea to run those utility in a controlled environment.
The process executor will run them in background, and we will execute
the process executor in detached mode.
Now that - the process executor runs in detached mode, we need an
observer, which will look at the status of the processes. It also reads
output, and error logs on demand.
Thanks - Surinder for helping in some of the UI changes.
into an error, and does not call the callback - 'success/fail'
internally.
And, that will not close the property dialog in those cases.
Reported by Harshal Dhumal.
required by its javascript module.
This will allow us to load the javascript modules as a static file, and
not as a Jinja2 template. This will increase the load time, as it will
decrease number of templates to be processed during loading those
javascripts.
will always have latest data related to that type of node. Also, fixed
the cache_level for different node types.
This commit also contains fixes for the following issue:
* In extension module - use the 'node-list-by-name' instead of using a
custom 'node-ajax-options' control, and removed redundant template
schemas from it.
* When we tries to destroy the select2 object from
Select2Cell/Select2Control while releasing the properties view,
sometimes select2 can not find the instance related it for some
unknown reason. Hence - before removing it we will check for manual
instance existance using $.data('select2').
* When we traverse through the browser tree nodes very quickly, it tries
to remove the object before it gets created completely, and results
into an exception.
* Icon in the select2 drop down list was not visible due to some CSS
issues.
Apart of that, we will generate two new browser events -
'pgadmin-node:created:<NODE-TYPE>', 'pgadmin-node:updated:<NODE-TYPE>'
whenever a new node is created, or an existing node will be updated.
the CAST module. Make the changes in the CAST module too for the
changes done in the NodeAjaxOptionsControl in the commit-id:
944bdbb96c.
We've changed the signature (arguments) of the 'options' & 'transform'
functions.
Also, remove the CodeMirror object for better clean up process.
When a SQL control/tab is rendered in a hidden element, it does not
render really well. Because - it rely on the height, and width of the
container to count the gutten position, and margin, and other elements.
Hence - whenever the tab, panel becomes visible, we will refresh the
control.
aciTree.
Also - fixed few CSS specifically for the preferences dialog.
Returning the sorted preferences from the server, so that - it lists
down in correct order.
due to internal 'set' function call from the internal 'reset' function.
We will stop the session management, before doing any clean up for
ignoring such events.
- Renaming or changing the schema for a shell type should not be allowed.
- I'm allowed to try to add ACL entries or security labels to an
existing shell type. This should be disallowed.
- Changing the schema on a (non-shell) type doesn't work - the type
name is omitted, e.g.
ALTER TYPE pem
SET SCHEMA pemhistory;
Which should be:
ALTER TYPE pem.foo
SET SCHEMA pemhistory;
on the object everytime, it hits the buttons, and the callback functions
are being executed. The original implementation was not using the same
panel, when it was recalled from other function. And, results into
unexpected behaviour like not able to work with Edit object again.
Thanks Murtuza for reporting the issue.
Removed the use of separate editor for both of these cell types. There
were two instance of select2 were getting created in the Select2Cell,
one in the Select2Cell itself, and another in Select2CellEditor. And,
loosing the focus mysteriously, and making the scrollbar in the property
dialog non-responsive.
Also, modified the NodeAjaxOptionsCell to use the above logic, and
removed its own version of render function to make it consitent across
the system.
This patch [changes sent by Murtuza] also includes improvisation in the
DeleteCell, and ObjectCell, which will honour now 'canRemoveRow', and
''canEditRow' respective properties of Column.
The current implementaton binds the cell/control object, and the ajax
data in the asychronous Cells/Controls with the 'options' functions
extended from the Select2Cell.
The problem starts when we try to fetch the current model from that
options/transform/filter function to do some operation, which does not
require in most of the cases. Except the privileges control - where we
needed the current model for omitting the existing selected object
during transformation, and filtering.
In order resolved the issue, we need a common object, which is shared
among the Cell. In backgrid, the 'Column' object is mong the cell,
hence - implementation logic has been changed to bid the 'Column' object
with the 'options' function and, passed the 'Cell' object as an
arguments.
Because - we do use the common function 'transform' between 'Control'
and 'Cell', we needed make changes in the Select2Control to pass the
Control object as an arguments.
And, make the changes in the privileges control to use the new
implementation. The same logic is also required in some of the
operations, we will be/are working on the table/column nodes.
operation.
Some of the nodes do not return id as result, when they were created.
Modified the logic to allow to open the parent node in such conditions.
multiple menu-items (not just the 'create' menu-group).
Moved all the main menu/context menu generation implementation in the
'menu' javascript menu.
In this implementation, if more than one menu-items specify same type
of categories, they will be created withing that group, otherwise - it
will be created separately (unless 'single' property of that category is
set to true).
We can also provide icon, priority, separator(s) above/below it for the
individual sub-menu too using pgAdmin.Browser.add_menu_category
function(...).
With previous implementation each node will have to send the values of 'show_system_objects'
while calling get_dependencies/get_dependents function. Now I have changed the logic to use the
value directly into the base class functions.
generated for the privilege control. Also, made changes in the
PrivilegeRoleModel, to allow to select a grantee only once for the
current user (as grantor), and omit them from other cell objects.
Valid/Invalid event was not properly triggered, whenever the child
attribute is a collection/model within a model (nested mode/collection
within model). This patch tried to take care of all such scenarios in
general.
change events.
- In input & textarea control, call the change event on key up (but -
not on every key up events, wait for sometime before trigger that
event, so that - we do not overdo that).
- In unique/subnode control, whenever we create new model object, set
all the required static fields used by pgAdmin form generation logic.
- In NodeAjaxListCell, we don't need to call the list generation
operation (ajax operation) for each and every cell. They use the same
shared column object.
- Remove unnecessary columns from collection property lists.
- Standardise property names and groupings.
- pg_attribute doesn't have an OID column, so use attnum instead (of atttypid!?!)
- Don't display the OID on Edit dialogues
- Use the ACL summary display in the properties panel, not the subnode editor.
- s/Oid/OID/g
- Ensure all security properties are displayed in the Security group
- Ensure all object definition values are stored in the Definition (or a more precise) group, not General.
- Fix the ordering of General properties so the comment is always show last.
- Display the ACL string in properties, not the subnode panel.
- Offer default values for the owner of new objects.
- Add missing context menu option to create a tablespace from a server node.
This will also allow us to operate on individual privileges & also we
needed this functionality for column nodes.
For example,
*Earlier:* priv was string
GRANT {{ priv }} ON {{ type }} TO {{ conn|qtIdent(role) }};
*Now:* priv will be List, which we need to handle in jinja templates.
GRANT *{{ priv|join(', ') }}* ON {{ type }} TO {{ conn|qtIdent(role) }};
which allows to determine certain group to hide/show the whole
tab/fieldset control in the properties, and create dialog.
Using this functionality in the database for the
'default privileges - types', which is not allowed before PG < 9.2 using
the version check.
templates. The variable scope for the jinja2 template variable is not
consitent across all the for loop (loop withing loop), it was behaving
inconsistently, a value set in a loop was not properly accessible in
inner loop properly. In order to resolve the issue - setting/using the
variable only in inner loop, and call add_menus function multiple times,
which we were trying to avoid, but - we can live with that.
(dynamic menu generation, typo in error message)
- Dynamic menu generation code was failing for some corner cases.
- Removed trailing whitespaces.
- Resolved a typo in the message - 'click_for_detailed_msg', which was
resulting in '{}' message instead of actual error message.
This patch also includes an issue related to IntegerControl, which does
not emit any error message, when it has some issues.
In order to fix the issue, it will trigger 'pgadmin-session:invalid'
event, when the input value is not an valid integer.
group, we do have the same function in CollectionModule class, which
will work, when we set the min_ver, max_ver and server_type properly.
Thanks Khushboo for pointing that out, and Akshay for quick testing for
this patch.
This commit takes care of the following issues/improvements.
* Adding missing imports for unauthorised in database module
* Node under Servers Nodes (i.e. Databases, tablespaces, roles nodes)
need to be inherited from PGChildNodeView (and, not from NodeView) for
adding server version check for their children.
* Adding statistics for database, and tablespaces in its node (not, yet
in UI)
* Renaming the camel case methods with proper name.
(i.e. getSQL -> get_sql, getNewSQL -> get_new_sql, etc.)
* Fixed the functions going beyond the text limit (column: 80) in
Databases, Roles & Tablespaces modules.
* Fixed the node method of Database module, which was not tested ever.
* We do not need separate SQL template for fetching the name (i.e.
get_name.sql), using the 'nodes.sql' for the same.
* Optimise the query for fetching ACLs for the database node, we didn't
require to join certain tables, while fetching only the ACLs.
* Introduced the list of the ACLs (regular and default ACLs for
different type) supported by each version [Databases Module].
* Renamed the templates 'get_nodes.sql' to' nodes.sql' to make it
consistent with other modules.
* Removed the checks for the authentication table use, as we don't need
to expose the password to the users (even the encrypted MD5). Using
the pg_roles view always for fetching roles/users information now.
* Resolved some typos in unreachable (specially the exceptions
catchment area.)
* Logging the exception in the application.
* Using qtLiteral, qtIdent properly in the templates (do not assume
about the types of data.)
* Using tsid as identifier instead of did for the tablespaces.
* Using nodes method of tablespace view for fetching individual node
information.
* Removing the hardcoded node information from the 'parse_priv_to_db'
function, and pass on allowed ACLs by the caller nodes.
* Using 'nodes.sql' to fetch name of the template instead of writing
that in the delete.sql template, which is definitely wrong place to
fetch the name of the object. [Tablespace Module]
called while initialing the CollectionNodeModule, as
super(..).__init__(..) does not call __init__ function of all parent
classes. Please refer to method resolution order (MRO) in python for
more information.
This patch also includes some typo resolution in PGNdoeModule. Also,
removed the AddAttr(..) at the moment, because - it is not going
anywhere at the moment.
