Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

Foreign Relations of the United States

The Foreign Relations of the United States (FRUS) series presents the official documentary historical record of major U.S. foreign policy decisions and significant diplomatic activity. The series is published in print and online editions at the U.S. Department of State Office of the Historian website.

In keeping with the spirit of the Digital Government Strategy's goals to provide the public with access to high-quality digital government information and services and help unlock the power of government data to spur innovation, the Office of the Historian is releasing the digital master source files for the online edition of the Foreign Relations series.

Organization and Format

The digital master source files for the Foreign Relations series have been prepared according to the Text Encoding Initiative (TEI) P5 Guidelines. Our project-specific encoding guidelines and conformance requirements can be found in the schema directory, where we have supplied a TEI ODD file, frus.odd, as well as Schematron and RelaxNG files used for additional conformance checks (see especially frus.sch, as frus.rng is generated from frus.odd). The volumes are stored in the volumes directory, one file per volume. Bibliographic information about all volumes in the series, including those volumes not yet available in full text TEI, is stored in the bibliography directory.

Locating the Source for a URL from

URLs for FRUS on all follow this pattern:{VOLUME_ID}/{ELEMENT_ID}

  • where {VOLUME_ID} is the volume identifier and corresponds to the base name of the files in the volumes directory.
  • and {ELEMENT_ID} is the identifier for an element within the volume's XML file and corresponds to the value of an @xml:id attribute on a TEI element, e.g., <div xml:id="ch1" type="chapter">.

Volume identifiers generally follow this hierarchical convention:

  • frus

For example, frus1969-76v19p1 is Volume XIX, Part 1 of the Nixon-Ford subseries (1969–76), within the Foreign Relations series.

Element identifiers generally fall into one of these types:

  • ch{NUMBER} or comp{NUMBER}: a chapter or compilation number
  • d{NUMBER}: a document number
  • d{NUMBER}fn{NUMBER}: a footnote within a document
  • p_{PERSON_ID} or t_{TERM_ID}: an entry for a person or term in the List of Persons or List of Terms & Abbreviations
  • {SECTION_ID}: a section title, e.g., preface corresponds to the Preface

For example, to locate the source data for, you should locate frus1969-76v24.xml in the volumes directory and search for a TEI element whose @xml:id attribute value is d176. You can safely assume that d176 is the <div> for Document 176 of this volume.

Not all information found at a given URL from necessarily comes from the element identifier in the URL. For example, in this Document 176, the left sidebar contains a table of contents for the entire volume (derived from the compilation and chapter <div> elements), and the right sidebar contains a listing of the people and terms & abbreviations used in the document (derived from the persons and terms glossaries in the front matter).

Canonical Identifiers

Once a volume has been released on and the HistoryAtState GitHub repository, the following identifiers are considered canonical and will not be altered:

  • Volume identifiers found in volume filenames and in the @xml:id attribute of each volume's root <TEI> element
  • Document identifiers found in the @xml:id attribute of each FRUS document's TEI <div> element

Very rarely, after we have digitized and released a volume in which documents were originally printed without document numbers, we may discover a problem in the way a document's boundary was encoded. To correct the boundary while respecting this policy, our practice is as follows:

  • Case 1: Document "d1" is discovered to contain 2 documents. Solution: The 2nd document is assigned a document identifier "d1a". No other documents' identifiers are altered.
  • Case 2: Document "d2" is discovered to belong to document "d1". Solution: The "d2" content is moved to "d1", but the "d2" div is not deleted; instead, placeholder text is supplied explaining what happened. An errata statement is appended to the volume too.

If any changes to document identifiers are made in the future, it will be to enhance their global uniqueness, e.g., to prepend the volume identifier to the document identifier, frus1969-76v18_d1. Any such changes will be announced via GitHub issue.

Other identifiers, such as person, and term identifiers (on <persName>, and <term> elements), are not yet considered canonical and are subject to alteration. At present, these identifiers are useful only for single-volume reference purposes. The Office plans to consolidate these per-volume identifiers into series-wide resources to aid in research and analysis.

Similarly, page break identifiers (on <pb> elements) are not considered canonical and will be corrected if errors are found.

Release Schedule

Updates to this repository are posted as soon as they are posted to Releases will include fixes to already published volumes, new publications, and digitized legacy volumes.


We welcome your feedback about this repository and how we can improve it. Please submit an issue or email us at


The files in this repository are in the public domain and may be copied and distributed without permission. For more information, see


  1. Single xar file: The collection.xconf will only contain the index, not any triggers!

  2. DEV environment: The replication triggers for the producer server are enabled in collection.xconf and point to the dev server's replication service IP.

    ant xar-dev
  3. PROD environment: Same as in 2. but for PROD destination

    ant xar-prod