Work Order Import

Clean up what a PDF parser reads with field filters

3 min read Work Order Import

A filter changes the value a PDF parser has read for one field before it is written. Use one to drop a label that came along with the value, to strip letters from an order number, or to put a prefix in front of a name. Filters belong to the parser as a whole, so they apply to every reference document.

Filters only run when a PDF is read. An email parser's text rules aren't filtered by this page. They have their own clean-up options. See Email parser rules.

Before you start

The page lists the fields that have a rule on a PDF parser. Until you have saved a reference document with at least one marked field, it says "Before you can filter fields, please upload a PDF and add some reference fields." and links to the parser page. Rules you have applied but not saved on the parser page don't appear here. See PDF parser rules.

Step by step

  1. Open your parser from Parse, then Field filters in the left menu.
  2. Find the field. Every field that has a saved rule has a block. A description marked with a Start and End label shows as one Project description block.
  3. Fill in any of the options below.
  4. Click Save.
Option What it does
Text to remove Type text and press enter. Add as many as you need.
Prepend text (added before value) Text put in front of the value, followed by a space.
Append text (added after value) Text put after the value, with a space before it.
Remove non-numeric characters Keeps only the digits.
Remove numeric characters Removes every digit.
Remove spaces Removes every space.

The changes sit on screen as you make them and are written when you click Save, which says "Parser filters saved."

Fields read as a block of contacts, custom fields and data points have no filter block here.

The order a filter works in

  1. Text to remove, one entry at a time.
  2. Remove non-numeric characters.
  3. Remove numeric characters.
  4. Remove spaces.
  5. Prepend text, then Append text.

A filter does nothing to a field the PDF gave no value for.

How Text to remove matches

An entry is removed everywhere it appears in the value, and capitals and spacing have to match what the PDF holds. "Job number -" removes those characters and leaves any space that followed them, so include the trailing space in the entry when you want it gone.

An entry with an asterisk removes whole words. "Ref*" removes every word that starts with "Ref", where a word is anything between spaces. The text before the asterisk is what the word has to start with.

When it goes wrong

A field I marked isn't listed

The page builds its list from the saved rules, so save the parser page first. A field that was marked as a data point or as a contact block isn't listed either.

The filter isn't applied to my email parser

Filters, like Field mapping, only run on a PDF. A PDF parser that an email parser links to uses its own filters.

Suggested rules added a filter that isn't saved

A suggested rule for a value that has its label in the same piece of text, such as "Job number - 16171", adds a filter that removes the label. It is held on screen with the rule, and only written when you click Save here. The parser page's Save keeps the rules and not the filters.

A prefix shows up in a description template

When a template inserts a data point, every filter on the parser is applied to the data point's text, whichever field the filter is for.