ProductOpener::Products - create and save products
ProductOpener::Products is used to create products and save them in Product Opener's database and file system.
use ProductOpener::Products qw/:all/;
my $product_ref = init_product($User_id, $Org_id, $code, $countryid);
$product_ref->{product_name_en} = "Chocolate cookies";
store_product("my-user", $product_ref, 'helpful comment');
When a product is saved, a new revision of the product is created. All revisions are saved in the file system:
products/[barcode path]/1.sto - first revision products/[barcode path]/2.sto - 2nd revision ... products/[barcode path]/product.sto - link to latest revision
The latest revision is stored in the products collection of the MongoDB database.
Before a product is saved, this module compute the completeness and quality of the data, and the edit history.
make_sure_numbers_are_stored_as_numbers() forces numbers contained in the product data to be stored as numbers (and not strings) in MongoDB.
Perl scalars are not typed, the internal type depends on the last operator used on the variable... e.g. if it is printed with a string concatenation, then it's converted to a string.
See https://metacpan.org/pod/JSON%3a%3aXS#PERL---JSON
assign_new_code() assigns a new unused code to store a new product that does not have a barcode.
my ($code, $product_id) = assign_new_code();
A list with the new code and the corresponding product_id.
This function currently assign new codes in sequence starting from 2000000000001. We increment the number by 1 for each product (which means codes are not valid as the last digit is supposed to be the check digit), and check if there is already a product for that number.
Codes starting with 2 are reserved for internal uses, there may be conflicts as other companies can use the same codes.
normalize_code() this function normalizes the product code by: - running the given code through normalization method provided by GS1 to format a GS1 data string, or data URL to a GTIN, - keeping only digits and removing spaces/dashes etc., - normalizing the length by adding leading zeroes or removing the leading zero (in case of 14 digit codes)
Product Code in the Raw form: $code
Normalized version of the code
On disk, we store product files and images in directories named after the product code, and we add leading 0s to the paths. So we need to normalize the number of leading 0s of product codes, so that we don't have 2 products for codes that differ only by leading 0s.
This function normalizes the product code by: - removing leading zeroes, - adding leading zeroes to have at least 13 digits, - removing leading zeroes for EAN8s to keep only 8 digits
Note: this function adds leading 0s even if the GS1 code is not valid.
normalize_code_with_gs1_ai() this function normalizes the product code by: - running the given code through normalization method provided by GS1 to format a GS1 data string, or data URI to a GTIN, - keeping only digits and removing spaces/dashes etc., - normalizing the length by adding leading zeroes or removing the leading zero (in case of 14 digit codes)
Product Code in the Raw form: $code
Normalized version of the code, and GS1 AI data string of the code, if a valid GS1 string was given as the argument
is_valid_code() checks if the given code is a valid product code.
Product Code: $code
Boolean value indicating if the code is valid or not.
split_code() this function splits the product code for determining the product path and the _id. product_path_from_id() utilizes this for the said purpose.
Product Code: $code
Code that has been split into 3 sections of three digits and one fourth section with the remaining digits. Example: 1234567890123 :- 123/456/789/0123
product_id_for_owner() returns the product id associated with a product barcode.
If the products on the server are public, the product id is equal to the product code.
If the products on the server are private (e.g. on the platform for producers), the product_id is of the form user-[user id]/[code] or org-[organization id]/code.
Product barcode
In most cases, pass $Owner_id which is initialized by ProductOpener::Users::init_user()
undef for public products user-[user id] or org-[organization id] for private products
The product id.
Returns the server for the product, if it is not on the current server.
undef is the product is on the current server, or server id of the server of the product otherwise.
Return the MongoDB database for the product: off, obf, opf, opff or off-pro
If we are on the producers platform, we currently have only one server: off-pro
Returns the relative path for the product.
Product id of the form [code], [owner-id]/[code]
The relative path for the product.
Returns the relative path for the product.
Product object reference.
The relative path for the product.
Reverse of product_path_from_id.
There is no guarantee the result will be correct... but it's way faster than loading the sto !
Initializes and return a $product_ref structure for a new product. If $countryid is defined and is not "en:world", then assign this country for the countries field. Otherwise, use the country associated with the ip address of the user.
Returns a $product_ref structure
Utility function to change the barcode of a product. Fails and returns an error if the code is invalid, or if there is already a product with the new code.
If successful: undef If there was an error: invalid_code or new_code_already_exists
Utility function to change the product type of a product. Fails and returns an error if the product type is invalid.
If successful: undef If there was an error: invalid_product_type
Compute sort keys that are stored in the MongoDB database and used to order results of queries.
Used on the web site for facets pages, except the index page.
Used for the Personal Search project to provide generic search results that apps can personalize later.
Save changes of a product: - in a new .sto file on the disk - in MongoDB (in the products collection, or products_obsolete collection if the product is obsolete)
Before saving, some field values are computed, and product history and completeness is computed.
Analyze the sources field of the product, as well as the changes to add to the data_sources field.
Sources allows to add some producers imports that were done before the producers platform was created.
The changes structure allows to add apps.
Function to do some normalization of product data (from the product database or input product data from a service)
For a specific change, analyze change identifiers (comment, user agent, userid etc.) to determine if the change was done through an app, the OFF userid, or an app specific UUID
reference to a change record
The function returns by order of preference: - a real user userid if we have an userid which is not the userid of an app - an appid + app uuid (e.g. some-app.Z626FZF4RTFSG6) - an app userid if the app did not provide an app uuid - openfoodfacts-contributors
For a specific product, replace a specific user_id associated with changes (edits, new photos etc.) by another user_id.
This can be used when we want to rename a user_id, or when an user asks its data to be deleted: we can rename it to a generic user account like openfoodfacts-contributors.
Find all products changed by a specific user_id, and replace the user_id associated with changes (edits, new photos etc.) by another user_id.
This can be used when we want to rename a user_id, or when an user asks its data to be deleted: we can rename it to a generic user account like openfoodfacts-contributors.
Record that a user has made a change of a specific type to the product.
For each type, there is a "list" array, and a "seen" hash
Returns a relative URL for a product on the website.
Returns a relative URL for an action on a product on the website.
This function is called by the web/panels/panel.tt.html template for knowledge panels that have associated actions.
Process the edit_rules (see @edit_rules in in Config file).
It applies in all API/form that edit the product. It applies to apply an image crop.
It does not block image upload.
Each condition is either a match on user_id or it's contrary user_id_not, or in_TAG_NAME_tags for a tag field to match a specific tag id.
Note that conditions are checked before editing the product !
ignore alone, ignore every edits.
You can also have rules of the form ignore_FIELD and warn_FIELD which will ignore (or notify) edits on the specific field.
Note that ignore rules also create a notification.
For nutriments use nutriments_NUTRIMENT_NAME for FIELD.
You can guard the rule on the field with a condition: ignore_if_CONDITION_FIELD or warn_if_CONDITION_FIELD This time it's to check the value the user want's to add.
CONDITION is one of the following:
Notifications are email addresses to send emails, or "slack_CHANNEL_NAME" (warning currently channel name is ignored, we post to edit-alerts)
Generates a text that describes the changes made. The text is displayed in the edit history of products.
$change_ref: reference to a change record
If the user who add or edits the product belongs to one or more teams, add them to the teams_tags array.
$product_ref
Checks if the product data should be protected from edits. e.g. official producer data that should not be changed by anonymous users through the API
Product data is protected if it has an owner and if the corresponding organization has the "protect data" checkbox checked.
- 1 if the data is protected - 0 if the data is not protected
Utility function to delete fields from a product_ref or a subfield.
Reference to a complete product a subfield.
An array of field names to remove.
Add fields like image_[front|ingredients|nutrition|packaging]_[url|small_url|thumb_url] to a product object.
If it exists, the image for the target language will be returned, otherwise we will return the image in the main language of the product.
Reference to a complete product a subfield.
2 language code of the preferred language for the product images.
Optional parameter to specify the type of image to add. Default is to add all types.
This function processes product raw data to analyze it and enrich it. For instance to analyze ingredients and compute scores such as Nutri-Score and Environmental-Score.
Reference to a product.
Reference to a response object to which we can add errors and warnings.
Return 1 if the field value was provided by the owner (producer) and the field is not a tag field.