MrmDumpType enumeration

Defines constants that specify the type of PRI file dump to produce when calling one of the MrmDumpPri... functions.

Syntax

typedef enum _MrmDumpType { 
  MrmDumpType_Basic     = 0,
  MrmDumpType_Detailed  = 1,
  MrmDumpType_Schema    = 2
} MrmDumpType;

Constants

MrmDumpType_Basic

Include string and file resources, but exclude embedded binary data. This is generally the most useful dump type for debugging.

MrmDumpType_Detailed

Include all resources, including embedded binary data, plus additional indexing information. This is most useful for round-tripping XML back to a PRI file or debugging advanced problems.

MrmDumpType_Schema

Include only resource names and indices, not candidates. This is the smallest dump type, and can be used when building versioned PRI files (see Remarks for more info).

Remarks

PRI files are stored an undocumented binary format, so the only way to view their contents is to dump them to XML via one of the MrmDumpPri... functions. Dumping the XML is useful for three purposes:

1. The XML is human-readable, enabling you to see the contents of the file, debug issues with unresolved resources or incorrectly-qualified candidates, and so on.
2. The XML is machine-readable, making it possible for tools to manipulate PRI files and generate new PRI files via the intermediate XML format.
   • Note that there is no built-in mechanism to turn an XML dump back into a PRI file; it must be done manually by creating a Resource Indexer, adding the resource candidates from the XML file, and writing out the new file.
3. The Schema dump type can be used to create versioned PRI files.

The Basic dump type is sufficient for most debugging purposes. It lists all resource names and candidate values, with the exception of embedded binary data - but since embedded binary data isn't human-readable anyway, that is typically not a problem.

The Detailed dump type is the most complete - including BASE64-encoded embedded binary data - but includes a lot of extra data that make it harder to read (especially if there are large embedded binary resources). A Detailed dump file can be used to round-trip back into the original PRI file.

The Schema dump type is the smallest dump type, and only includes information about the resource names and indices in the PRI file (the "Schema" of the file). A Schema dump can be used with one of the MrmCreateResourceIndexerFromPreviousSchema... functions to create versionable PRI files - see the Remarks in MrmCreateResourceIndexerFromPreviousSchemaFile for more info. Note there is no guarantee that the resources listed in a Schema dump actually have candidates defined in the PRI file, so this is insufficient for debugging "missing" resource errors.

Requirements

See also

MrmDumpPriDataInMemory

MrmDumpPriFile

MrmDumpPriFileInMemory

Package resource indexing (PRI) APIs and custom build systems