---
name: "Mail::SpamAssassin::Plugin::MIMEHeader"
section: "3pm"
platform: "linux"
package: "spamassassin"
version: "4.0.2-3"
release: "Arch Linux 2026-11-01"
platforms: ["linux"]
url: "https://new.carta.tech/Mail::SpamAssassin::Plugin::MIMEHeader"
---

# Mail::SpamAssassin::Plugin::MIMEHeader(3pm)

perform regexp tests against MIME headers

## NAME

MIMEHeader - perform regexp tests against MIME headers

## SYNOPSIS

```
  loadplugin    Mail::SpamAssassin::Plugin::MIMEHeader
  mimeheader    NAME_OF_RULE    Content-Id =~ /foo/
```

## DESCRIPTION

This plugin allows regexp rules to be written against MIME headers in the message.

## RULE DEFINITIONS AND PRIVILEGED SETTINGS

- mimeheader NAME\_OF\_RULE Header-Name =~ /pattern/modifiers

  Specify a rule.  `"NAME_OF_RULE"` is the name of the rule to be used, `"Header-Name"` is the name of the MIME header to check, and `"/pattern/modifiers"` is the Perl regular expression to match against this.

  Note that in a message of multiple parts, each header will be checked against the pattern separately.  In other words, if multiple parts have a 'Content-Type' header, each header's value will be tested individually as a separate string.

  Header names are considered case-insensitive.

  The header values are normally cleaned up a little; for example, whitespace around the newline character in "folded" headers will be replaced with a single space.  Append `":raw"` to the header name to retrieve the raw, undecoded value, including pristine whitespace, instead.

- tflags NAME\_OF\_RULE range=x-y

  Match only from specific MIME parts, indexed in the order they are parsed. Part 1 = main message headers. Part 2 = next part etc.

  ```
   range=1    (match only main headers, not any subparts)
   range=2-   (match any subparts, but not the main headers)
   range=-3   (match only first three parts, including main headers)
   range=2-3  (match only first two subparts)
  ```

- tflags NAME\_OF\_RULE concat

  Concatenate all headers from all mime parts (possible range applied) into a single string for matching.  This allows matching headers across multiple parts with single regex.  Normally pattern is tested individually for different mime parts.