AuthoringMemory

Owned by Claudius Hörth

Feb. 08, 2024

POST

/congree/api/AuthoringMemory/v1/ImportXml?lang={CultureName}

HTTP request used to add content to the Congree Sentence Database

The culture name in the format languagecode2-country/regioncode2. languagecode2 is a lowercase two-letter code derived from ISO 639-1. country/regioncode2 is derived from ISO 3166 and usually consists of two uppercase letters, or a BCP-47 language tag.

E.g. /congree/api/AuthoringMemory/v1/ImportXml?lang=de-DE

The content type should be multipart/form-data and it should contain file XML.

Response

{
  "DetectedParagraphsCount": 0,
  "DetectedSentencesCount": 0,
  "AddedSentencesCount": 0,
  "FailedParagraphsCount": 0
}

Requirements

The content has to be well-formed XML with resolved entity references.

Maximum size: 10 MB

Authoring Memory: Search

POST

/api/AuthoringMemory/v1/Search

Searches for sentence in Authoring Memory

This endpoint requires a JSON object as parameter:

{
  "Xml": "string",
  "RuleSetName": "string"
}

Property	Description
Xml	A well-formed XML string.
RuleSetName	The name of the rule set the string is assigned to.

The request returns a JSON object providing the information detailed below.

{
  "ResultXml": "string",
  "Sentences": [
    {
      "Id": 0,
      "ContextId": 0,
      "DisplayContent": "string",
      "SearchResults": [
        {
          "SearchMode": "Similar",
          "Similarity": 0,
          "Class": "string",
          "DisplayContent": "string",
          "ApplyContent": "string",
          "Evaluation": "Approved",
          "Attributes": [
            {
              "Name": "string",
              "Values": [
                "string"
              ]
            }
          ],
          "Penalties": [
            {
              "PenaltyType": "DifferentText",
              "Value": 0
            }
          ]
        }
      ]
    }
  ]
}

Property	Description
ResultXml	The string that was searched for annotated with processing instructions (PI) providing a markup to identify indiviual portions of the text.
Sentences	An array of all sentences that were found
Id	Id - identifier of original sentence. It refers to corresponding PI in ResultXml
ContextId	Identifier of paragraph. It refers to the corresponding PI in ResultXml.
SearchResults	An array of search results for a specific sentence.
SearchMode	The search method that was applied to get this result. Possible values are "Similar", "Concordance", "Word", "Approximate".
Similarity	The computed similarity between two sentences. This value is calculated if SearchMode is Similar.
Class	Congree class
DisplayContent	Contains the content of the found sentence. It can also contains the additional Processing Instructions which Congree adds to mark the difference between the original and the found sentences, where <?cngr-dif-b TypeOfDifference?> - indicates the beginning of the range where the difference in the content begins <?cngr-dif-e TypeOfDifference?> - indicates the end of the range where the difference in content ends <?cngr-dif-r TypeOfDifference?> - indicates that the content was removed at this location Each Processing Instruction contains as a PIContent an integer value that indicates the type of difference (TypeOfDifference). The the type of difference can take the following values: 2 - Authoring Memory content contains more characters than checked content. 3 - Checked content contains more characters than the Authoring Memory. 4 - Indicates that Auto-Adjustment was applied at this point. 5 - Indicates that "placeholder" was detected. Congree recognizes tag as placeholder when it is defined as "Ignore content" in the paragraph recognition Treatments. How "placeholder" works: E.g. Let's assume <u> is a placeholder. If an original sentence is: This is a test <u>certain content that should not be considered when searching</u>. And found result is This </u> is not a test. Then the resulting sentence will be This <u>certain content that should not be considered when searching</u> is not a test.
ApplyContent	The content that Congree proposes to insert into the editor.
Evaluation	The Data Grooming state. Possible values are "Approved", "Discarded", "Unknown".
Attributes	An array of attributes associated with found sentence,
Name	The name of ther attribute
Values	An array of values assigned to the attribute
Penalties	An array of penalties imposed for mismatching the original proposal.
PenaltyType	The type of penalty. Supported types are: DifferentText, DifferentInlineElements, DifferentPunctuation, DifferentSpecialCharacters, DifferentClass, SwitchedWords, DifferentCase, DifferentSpaces, DifferentNumbers.
Value	The value deducted from the similarity after the fine is imposed

Authoring Memory: Attributes

POST	/api/AuthoringMemory/v1/Attributes	Adds new attributes with values
GET	/api/AuthoringMemory/v1/Attributes	Returns available attributes
DELETED	/api/AuthoringMemory/v1/Attributes	Deleted available attributes

Attributes can be specified by:

Type (string, optional): Type = ['Text', 'Picklist'],
Name (string, optional): Name
Values (Array[string], optional): Available values

Authoring Memory: Sentences

POST	/api/AuthoringMemory/v1/Sentences/{sentenceId}/Attributes	Adds new attibutes or values to the sentence
GET	/api/AuthoringMemory/v1/Sentences/{sentenceId}/Attributes	Gets attributes for certain sentence
DELETE	/api/AuthoringMemory/v1/Sentences/{sentenceId}/Attributes	Deletes attributes or values of the sentence
GET	/congree/api/AuthoringMemory/v1/Sentences	Get sentences based on the provided filter
DELETE	/congree/api/AuthoringMemory/v1/Sentences	Cleans up Authoring Memory sentences
DELETE	/congree/api/AuthoringMemory/v2/Sentences	Deletes sentences from Authoring Memory

The body of the request should contain the options you can find in the table below in the following form:

Response

{
  "ClearImportedSentences" : true,
  "ClearDataGroomingSentences": true
}

Option name	Description
ClearImportedSentences	Deletes imported (via API /congree/api/AuthoringMemory/v1/ImportXml) sentences
ClearDataGroomingSentences	Deletes all sentences from Congree Data Grooming and also related sentences from the Authoring Memory

In order to get the task status and results, use the methods described in Tasks. The result is an empty object.

Response

{
  "Id": "00000000-0000-0000-0000-000000000000"
}

Renponse example

{
  "Id": "514290f4-fc30-48e3-99b8-0973011bd2e5"
}