Creating Top Texts for an Object List

Use

If you create top texts for an object list, the system scans the source texts in the object list. It then creates top text objects that contain the source texts matching your selection criteria. The source texts are sorted according to the frequency with which they occur in the objects in the object list.

Translators process these top text objects, and should create status A proposals for as many unambiguous source texts as possible. You can then run automatic distribution to distribute these proposals and reduce the manual translation workload.

Prerequisites

To create top texts, you need the following authorizations:
- Authorization object S_ADMI_FCD with value TRNR .
- Translator profile XX Super User for all target languages for which you want to create top texts.
You must create an object list that only contains the objects relevant for translation into the target languages that will process these top texts. If it contains objects that are not relevant for translation into these target languages, the translators will waste time creating proposals for source texts that do not require translation in the first place. Either create a new object list or edit an existing object list . If you want to create top texts on the basis of more than one object list, you need to merge them first.

Example
Your object list contains objects of object type CLAS, even though you defined this object type as not relevant for translation into any target language in LXE_MASTER. If you use this object list as the basis for top text creation, the translators will create proposals via the top texts for CLAS objects, even though these objects are not translation-relevant.

If your object list contains objects from object types or collections that are not relevant for translation, it would be better to create top texts according to each target language's worklist. This method takes account of all the LXE_MASTER settings regarding relevance of object types, collections, and graphs for each target language. For more information, see Creating Top Texts for a Worklist .
We recommend that you evaluate your object list before creating top texts. You can then see how many lines were translated via distribution after top texts were processed, by comparing the "before" statistics with the "after" statistics. For more information, see Creating an Evaluation Run .

Procedure

Call up transaction LXE_MASTER and choose DEMS Objects Top Texts According to Object List .

Make the following entries in the Create Top Texts According to Object List screen:

Source language of texts in top text objects	In the Source Language field, enter the language in which the source texts should appear in the top text objects. Example If you are creating top texts for target languages that translate from source language German, enter deDE in this field. If you have target languages that translate from German and target languages that translate from English, you must schedule this job twice to create two sets of top text objects, once for deDE and once for enUS
Object list to be scanned	Choose F4 on the Object List field and in the dialog box that appears, select the object list that should be the basis for the top texts.
Original language of texts to be scanned	In the Original Language field, enter the original language of the relevant objects in the object list, if you want to ensure that the top text objects do not contain source texts which are actually translations of the original texts. Example You develop in both German and English, and you are creating top texts for target language English. If you leave this field blank, then the top texts could contain German texts that are actually translations of texts created in English that do not require translation into English. You must enter deDE in the Original Language field to ensure that the top texts only contain source texts requiring translation into English.
Name of top texts	Enter a text string in the Object Name field that will help you recognize the top texts when you call them up in transaction SE63.
Description of top texts	You can enter a description of up to 40 characters in the Description field to document how you created the top texts, for example.
Frequency of source texts	Enter the minimum repetition level of the source texts in the Minimum Frequency field. Example If you only want the top texts to contain source texts that occur 5 times or more in the object list, enter 5 in this field.
Target language	If you want to create the top texts for one target language only, and make use of the settings in the next five fields (see next 2 rows of this table), enter the target language in the Target Language field. To create top texts for several target languages, leave the default setting of * in this field and do not change the default settings in the next five fields.
Status of proposals	These settings are only possible if you want to create top texts for one target language and have entered the target language in the Target Language field. Here you can restrict the source texts that the top texts contain according to whether proposals exist for the source texts yet. Example To create top texts that only contain source texts for which no proposals exist as yet, leave the Minimum Status field blank and enter 1 in the Maximum Status field. To create top texts that only contain source texts for which status B proposals exist (to check the work of beginner translators, for example), choose F4 on the Minimum Status field and select status B , then choose F4 on the Maximum Status field and select status B again. The system enters the lower and upper numeric values of quality status B respectively. To create top texts that only contain source texts for which proposals of status B or S exist (to find any proposals that could be upgraded to status A), choose F4 on the Minimum Status field and select status B , then choose F4 on the Maximum Status field and select status S .
Status of texts to be scanned	These settings are only possible if you want to create top texts for one target language and have entered the target language in the Target Language field. As long as you have evaluated the object list, you can also restrict the source texts contained in the top texts according to the translation status of the lines in which the source texts occur. Example To create top texts that only contain source texts that occur in new lines (where the best distribution potential exists), ensure that the New Texts field contains the value X , and the Modified Texts and Translated Texts fields are blank. To create top texts that only contain source texts that occur in new and modified lines, ensure that the New Texts and Modified Texts fields contain the value X , and the Translated Texts field is blank.
Split top text objects according to domain	If your object list contains objects from collections belonging to different domains, we recommend that you create your top texts according to domain. If you leave the field Ignore Application blank, the system creates domain-specific top texts. For example, the system creates separate top text objects for the source texts in objects from Basis (domain BC) that match your selection criteria, and so on. This allows translators to create application standards for the domain in question; in cases where it is not possible to create a system standard for the source text, but an application standard for domain BC is perfectly possible.
Split top text objects according to object type	If you have already decided to split the top texts according to domain, we recommend that you leave the value X entered in the Ignore Object Type field; otherwise the system will create a large number of top text objects where the same source texts occur repeatedly. However, you may want to split the top texts according to object type to see the object types in which the source texts occur. If so, leave the Ignore Object Type field blank and enter an X in the Ignore Application field.
Include similar texts	If you leave the Ignore Similar Texts field blank, then the system scans for similar texts and groups them together in the top text objects (for example, Create Document , Create document , Create Document 2 ). This enables you to create consistent translations in the proposal pool. The similar texts listed in the top texts do not necessarily match the other criteria you defined for top text creation. For example, the top texts could contain similar texts for which proposals already exist, even though your top text creation criteria specified otherwise. If you enter the value X in the Ignore Similar Texts field, no scan for similar texts takes place.
Generate object list	If you enter the value X in the Generate Object List field, the system creates an object list that contains top texts created in the run. This enables you to create worklists for the top texts by evaluating this object list. To do this, you need to create a graph for collection SAP_TRANSLATION_DEMS and then activate it for the target languages that will translate the top texts. However, you are not obliged to create worklists for top texts. For other ways of accessing the top texts for processing, see Processing Top Texts .
Schedule top text job in the background	Ensure the Start in Background field contains the value X so that the job to create the top texts runs in the background.

Choose (F8) and schedule as a background job.

Result

As long as source texts matching your selection criteria exist in the object list, the system creates top text objects containing the appropriate source texts, sorted according tohow frequently they appear in the objects in your object list.

The spool displays the names of the top texts created by this job. The names of the top texts are structured as follows:

So we can learn the following from the object name deDE-TOP__C-INITIAL_FR-frFR-R-BC 0002 :

It contains German (deDE) source texts
It was given the name INITIAL_FR when the top text job was scheduled
It contains source texts extracted from objects belonging to the Basis (BC) domain
It is the second top text object created by this job (0002)

You can now trigger the processing of top texts .