Advancing open source analysis and development
|
FOSSology Advancing open source analysis and development |
Buckets allow a fossology system administrator to create reporting categories. For example, HP uses categories like “Preferred Licenses”, “SHIP-HOLD Licenses”, “Possibly restrictive licensing terms”, “Commercial Licenses”, … See the example report Buckets.
You don't have to use buckets if you don't want, but they are a very powerful way to organize your reports, minimizing the time needed for software reviews. Since buckets frequently need to make a value judgement, they need to be customized for each user/project/corporation. Samples (TBD) will ship with fossology.
Bucket analysis is done through a bucket agent and results are stored in the database.
Before you create a definition is is important to understand that:
These are the methods you can use to define a bucket:
Any item that doesn't match a defined bucket goes into a special bucket that just says it's not in any other one.
Buckets are individual classifications e.g. “Items with bad licenses”. But buckets are usually (always?) used in groups. Bucket Pools are collections of buckets. For example, you could have a bucket pool for your distro, project or company that defines good, bad, questionable, … buckets.
For every bucket pool, there is always a bucket called “No Match”. Any pfile that matches no other bucket will end up in this bucket.
Table: bucketpool
Names the each bucket pool.
| Column | Description |
|---|---|
| bucketpool_pk | Primary Key |
| bucketpool_name | Name of bucket pool, unique |
| bucketpool_desc | Bucket pool description |
Table: bucket
Defines each bucket in a bucket pool. Each bucketpool has 1-n bucket records.
| Column | Description |
|---|---|
| bucket_pk | Primary Key |
| bucketpool_fk | Foreign key to bucketpool this bucket definition belongs in. |
| bucket_name | Name |
| bucket_color | Color to highlight bucket in reports |
| bucket_report_order | Order to sort buckets in reports. 1 is highest priority |
| bucket_eval_order | Order to evaluate buckets. e.g. an order of 5 will be evaluated before an order of 6 |
| bucket_exclusive | True if an item that matches this bucket def will appear exclusively in this bucket. Note the eval_order is important. |
| bucket_type | 1=MATCH_EVERY, 2=MATCH_ONLY, 3=REGEX, 4=EXEC |
| bucket_regex | if type is REGEX, this is the regular expression |
| bucket_filename | if type is 1 or 2, this is the name of the file containing the license name strings. If type is 4, this is the name of the file to exec. Files are stored in $DATADIR/{bucketpool name} |
Table: bucket_file
Associate a pfile with 1-n buckets.
| Column | Description |
|---|---|
| bf_pk | Primary Key |
| bucket_fk | Foreign key to table bucket |
| pfile_fk | Foreign key to table pfile |
Buckets are based on an internal HP tool called Nomos. Here is a partial screenshot:
