# CSV Input Filtering - [Python Regular Expressions](Python-Regular-Expressions) Search function - [Definitions](#definitions) - [Quoting rules](#quoting-rules) - [Column row filtering](#column-row-filtering) - [Column row limiting](#column-row-limiting) - [Saving filters in gam.cfg](#saving-filters-in-gamcfg) - [Validate filters](#validate-filters) There are two values in `gam.cfg` that can be used to filter the input from `gam csv` commands. * `csv_input_row_filter` - A list or JSON dictionary used to include specific rows based on column values * `csv_input_row_drop_filter` - A list or JSON dictionary used to exclude specific rows based on column values These filters can be used alone or in conjunction with the `matchfield|skipfield ` options. * https://github.com/taers232c/GAMADV-XTD3/wiki/Bulk-Processing#csv-files ## Definitions [Data Selectors](Collections-of-items) ``` ::= | | ``` ``` ::= -- | (+|-)(d|w|y) | never| today ::= --T::[.](Z|(+|-(:))) | (+|-)(m|h|d|w|y) | never| now|today ::= <|<=|>=|>|=|!= ::= See: https://docs.python.org/3/library/re.html> :: = ::= [(any|all):]count| [(any|all):]countrange=/| [(any|all):]countrange!=/| [(any|all):]date| [(any|all):]daterange=/| [(any|all):]daterange!=/| [(any|all):]length| [(any|all):]lengthrange=/| [(any|all):]lengthrange!=/| [(any|all):]text| [(any|all):]textrange=/| [(any|all):]textrange!=/| [(any|all):]time| [(any|all):]timerange=/| [(any|all):]timerange!=/| [(any|all):]boolean:| [(any|all):]regex:| [(any|all):]regexcs:| [(any|all):]notregex:| [(any|all):]notregexcs:| [(any|all):]data:| [(any|all):]notdata:| ::= "':'(,':')*" ::= '{"": ""(,"": "")*}' | "{\"\": \"\"(,\"\": \"\")*}" ``` ## Quoting rules Name:value form. ``` ::= "':'(,':')*" ``` * ``, even if it has one element, should be enclosed in `"`. * Each `:` pair should be enclosed in `'`. * If `` contains a `:` or a space, it should be enclosed in `\"`. * If `` or `` in `` contain a space, it should be enclosed in `\"`. Example: ``` csv_input_row_filter "'\"accounts:used_quota_in_mb\":count>15000'" csv_input_row_filter "'email:data:\"csvfile gsheet:email user@domain.com FileID Sheet1\"'" ``` JSON form. ``` ::= '{"": ""(,"": "")*}' | "{\"\": \"\"(,\"\": \"\")*}" ``` * The first JSON form can be used on Linux and Mac OS; it can not be used on Windows. * The second JSON form can be used on Linux, Mac OS and Windows. * If `` contains a `:` or a space, no additional quoting is required Example: ``` csv_input_row_filter '{"accounts:used_quota_in_mb": "count>=150"}' csv_input_row_filter "{\"accounts:used_quota_in_mb\": \"count>=150\"}" ``` ## Column row filtering Row filtering includes/excludes rows based on column values. ### Field names Field names are specified by regular expressions; at its simplest, you specify a complete field name. Field names are matched in a case insensitive manner. If the field name doesn't contain any of the following regular expression characters `^$*+|$[{(`, it will be surrounded with `^$` so that it doesn't match any subfields that begin with the field name as a prefix. The following filter will match the count field and not the subfields. ``` config csv_input_row_filter "'externalIds:countrange=1/10'" primaryEmail,externalIds,externalIds.0.type,externalIds.0.value,externalIds.1.type,externalIds.1.value,... ``` ### Inclusive filters You can include rows for gam csv commands based on column values. You specify a list of fields(headers) and the values they must have. `csv_input_row_filter` is used to specify the fields and values. Each field name/expression can appear only once in the list. You specify whether all or any value filters must match for the row to be included in the input. * `csv_input_row_filter_mode allmatch` - All value filters must match for the row to be included in the input; this is the default * `csv_input_row_filter_mode anymatch` - Any value filter must match for the row to be included in the input ``` gam config csv_input_row_filter ... gam config csv_input_row_filter ... ``` ### Exclusive filters You can exclude rows for gam csv commands based on column values. You specify a list of fields(headers) and the values they must not have. `csv_input_row_drop_filter` is used to specify the fields and values. Each field name/expression can appear only once in the list. You specify whether all or any value filters must match for the row to be excluded from the input. * `csv_input_row_filter_drop_mode allmatch` - If all value filters match, the row is excluded from the input * `csv_input_row_filter_drop_mode anymatch` - If any value filter matches, the row is excluded from the input; this is the default ``` gam config csv_input_row_drop_filter ... gam config csv_input_row_drop_filter ... ``` ### Matches A filter matches if the field has the desired value. lf you specify a regular expression for a field name that matches several columns, the filter matches if any of the columns has a match. In the case of `notregex|notregexcs|notdata`, the filter matches if none (not any) of the columns has a match. `` allows specifying that the filter will match only if all of the columns have a match. In the case of `notregex|notregexcs|notdata`, the filter matches if some (not all) of the columns have a match. If neither `any` or `all` is explicitly specified, `any` is the default. These are the row value filter types: * `count` - Used on fields with numbers; a blank field will not match * `countrange=/` - Used on fields with numbers; a blank field will not match * The field value must be `>=` the left `` and `<=` the right `` * `countrange!=/` - Used on fields with numbers; a blank field will not match * The field value must be `<` the left `` or `>` the right `` * `date` - Used on fields with dates or times; only the date portion of a time field is compared; a blank field will not match * `daterange=/` - Used on fields with dates or times; only the date portion of a time field is compared; a blank field will not match * The field value must be `>=` the left `` and `<=` the right `` * `daterange!=/` - Used on fields with dates or times; only the date portion of a time field is compared; a blank field will not match * The field value must be `<` the left `` or `>` the right `` * `length` - Used on fields with strings; non string fields will not match * `lengthrange=/` - Used on fields with strings; non string fields will not match * The field length must be `>=` the left `` and `<=` the right `` * `lengthrange!=/` - Used on fields with strings; non string fields will not match * The field length must be `<` the left `` or `>` the right `` * `text` - Used on fields with text * `textrange=/` - Used on fields with strings * The field value must be `>=` the left `` and `<=` the right `` * `textrange!=/` - Used on fields with strings * The field value must be `<` the left `` or `>` the right `` * `time` - Used on fields with times; a blank field will not match * `timerange=/` - Used on fields with times; a blank field will not match * The field value must be `>=` the left `` and `<=` the right `` * `timerange!=/` - Used on fields with times; a blank field will not match * The field value must be `<` the left `` or `>` the right `` * `boolean:` - Used on fields with Boolean values; a blank field is considered False * `regex:` - Used on fields with text; field value must match ``; case insensitive * `regexcs:` - Used on fields with text; field value must match ``; case sensitive * `notregex:` - Used on fields with text; field value must not match ``; case insensitive * `notregexcs:` - Used on fields with text; field value must not match ``; case sensitive * `data:` - Used on fields with text; field value must match some value in ``; case sensitive * `notdata:` - Used on fields with text; field value must not match any value in ``; case sensitive ### **Change in behavior.** In versions prior to `5.12.00`, `regex:` and `notregex:` were processed in a case sensitive manner; in many cases this is probably not desirable; e.g., matching file names which are case insensitive. Now, `regex:` and `notregex:` are processed in a case insensitive manner. To get the prior case sensitive processing, use `regexcs:` and `notregexcs:`. ### Examples You want to process groups with 100 or more direct members. ``` gam redirect csv GroupInfo.csv print groups fields directmemberscount gam config csv_input_row_filter "'directMembersCount:count>100'" csv GroupInfo.csv gam group "~email" ... ``` You want to process groups not created by an administrator. ``` gam redirect csv GroupInfo.csv print groups fields admincreated gam config csv_input_row_drop_filter "'adminCreated:boolean:true'" csv GroupInfo.csv gam group "~email" ... ``` You want to process users created in the last 30 days. ``` gam redirect csv UserInfo.csv print users fields creationtime gam config csv_input_row_filter "'creationTime:date>=-30d'" csv UserInfo.csv gam user "~primaryEmail" ... ``` You want to process users that are consuming more than 15GB of storage. Special quoting is required because the field name contains a colon. ``` gam redirect csv UserInfo.csv report user services accounts fields "accounts:used_quota_in_mb" gam config csv_input_row_filter "'\"accounts:used_quota_in_mb\":count>15000'" csv UserInfo.csv gam user "~primaryEmail" ... ``` ## Column row limiting You can limit the number of rows read from a CSV file. You want to process the first 10 users that are consuming more than 15GB of storage. Special quoting is required because the field name contains a colon. ``` gam redirect csv UserInfo.csv report user services accounts fields "accounts:used_quota_in_mb" gam config csv_input_row_filter "'\"accounts:used_quota_in_mb\":count>15000'" csv_input_row_limit 10 csv UserInfo.csv gam user "~primaryEmail" ... ``` ## Saving filters in gam.cfg If you define a value for `csv_input_row_filter`, `csv_input_row_drop_filter` or `csv_input_row_limit` in the `[DEFAULT]` section of `gam.cfg`, it will apply to every `gam csv` command which is probably not desirable. You can store them in `gam.cfg` in named sections. ``` [Filter510] csv_input_row_filter = 'phones.\\\d+.value:regex:(?:^\\$510\\$ )|(?:^510[- ])\\\d{3}-\\\d{4}' ``` You want to process users with phone numbers in the area code 510; the number can be in the format `(510) ddd-dddd` or `510-ddd-dddd` or `510 ddd-dddd`. ``` gam redirect csv UserInfo.csv print users fields name,phones gam selectinputfilter Filter510 csv UserInfo.csv gam user "~primaryEmail" ... ``` ## Validate filters Version `6.30.00` added the `gam comment *` command that can be used to validate input row filters. ``` $ more Comment.csv col1,col2 aaa,111 bbb,222 ccc,333 $ gam config csv_input_row_drop_filter "col1:regex:bbb" csv Comment.csv gam comment "Col1:~~col1~~" "Col2:~~col2~~" 2022-12-16T12:41:50.045-08:00,0/2,Using 2 processes... Col1:aaa Col2:111 Col1:ccc Col2:333 $ gam config csv_input_row_filter "col1:regex:bbb" csv Comment.csv gam comment "Col1:~~col1~~" "Col2:~~col2~~" 2022-12-18T09:42:26.108-08:00,0/1,Using 1 process... Col1:bbb Col2:222 ```