Optimizations and CSV Import Rework
- Replaced " with ' where possible to prevent the parser from being called when not needed.
- Replaced == and != with === and !== where possible for bug prevention and speed.
- Replaced -1 with NEW_ITEM global constant for code clarity.
- Added NEW_ITEM global constant to constants.php.
- Refactored CSV import function names for clarity.
- Added capability to import a CSV file containing updates.
- Replaced array() with [] for speed and consistency.
- Removed hungarian notation from two private functions.
- Refactored QueryBuilder functions to place table name in the get() function call.
- Replaced (int) cast with call to intval() for speed.
- Replaced == and != with === and !== where possible to prevent bugs and for speed.
- Replaced array() with [] for speed and consistency.
- Fixed search_custom call
Optimizations and bugfixes for attributes used in csv_import
- Reordered where statements in queries to match composite index on attribute_links table.
- fixed value_exists() to account for different attribute types.
- Removed hungarian notation on private function.
- Replaced array() with [] for speed and consistency.
- Replaced != with <> in SQL for consistency.
- Removed from() calls in querybuilder where possible to reduce function calls.
- Add get_items_by_value()
- Reworked check_data_validity()
- Remove unneeded comments
- Refactor functions for code clarity.
- Use $this->db->dbprefix() where possible instead of hand-writing ospos_...
- Removed unneeded column from query.
- Replaced (int) cast with intval() call for speed.
- Added get_attribute_values()
- Fixed issue with date format locale not being used
- Refactored save_value to respect different attribute_types
- Added delete_orphaned_links() to remove attribute_links that are no longer linked to any items
- Added get_attributes_by_definition()
- Added attribute_cleanup()
Optimizations used in csv_import
- replaced array() with [] for consistency and speed.
- Removed hungarian notation in private functions.
- Replaced " with ' where possible to prevent the parser from being called.
- Minor formatting
- Refactored if statement to tertiary notation for cleaner implementation.
- Replaced " for ' where possible to prevent the parser from being called.
- Added the Id column in the generate_import_items_csv() template so that users can submit an update to an existing item.
- Removed unused key=>value pairs in foreach loops for speed.
- Removed unneeded comments where the function name was self-explanatory.
- Rework get_csv_file() for speed.
- Rework bom_exists() for speed.
- Replaced array() with [] for speed and consistency.
- Replaced == with === where possible to prevent bugs and for speed.
- Reworked valid_date() and valid_decimal helper functions for speed and accuracy according to the locale_format instead of a fixed format.
- Minor Reformatting for clarity.
- Replaced " for ' to prevent the parser from being called.
- Refactored function call names to reflect new names.
- Added missing ; in
- Used String interpolation where useful.
- Spelling fix in comment
Requested Review Changes
- Fixed indentation in Items.php
- Fixed indentation in Attribute.php
- Refactored variable out of long line of code to make it more readable.
- Removed unnecessary indices on deleted column where there would not be many rows.
- Removed unnecessary parenthesis in Item Model.
- Added Composite indices based on MySQL EXPLAIN results.
Set group_concat_max_len session variable to prevent count totals at the bottom of the tables becoming `1 to null of null items` when database includes significant number of item rows.
per #2788 the Supplier table results from Item->get_all() and Item->get_info() are not used anywhere and causing a bug where the duplicate fields between the items table and the suppliers table (deleted and category) are returning just one value to the results. This creates the undesired effect of expecting items.deleted and items.category but getting values from suppliers.deleted and suppliers.category.
