Glossary and Translation Memory File Formats
Lectio imports standard translation-memory and terminology exports from Smartcat, Trados, MultiTerm, memoQ, Phrase, and other CAT tools. Use the Import action on the Glossaries or Translation Memories page to open the migration workspace.
Each source file can be up to 500 MB. A migration can contain up to 10 files and 2 GB of source data.
Translation Memory Imports#
| Format | Extension | What Lectio imports |
|---|---|---|
| Translation Memory eXchange | .tmx | Multilingual translation units, language codes, context, notes, properties, dates, identifiers, and supported inline content |
| Excel workbook | .xlsx | One source column, one or more target-language columns, and optional context or metadata columns |
| Delimited text | .csv, .tsv | The same column mapping as Excel, with quoted values and header rows |
TMX is the best choice when your CAT tool offers it. It preserves language and translation-unit metadata without requiring manual column mapping.
When the same source text has different target translations, Lectio shows the conflict count and samples before import. You can keep the first target or explicitly keep the last target. Exact duplicates are skipped.
Glossary Imports#
| Format | Extension | What Lectio imports |
|---|---|---|
| TermBase eXchange | .tbx, .xml | TBX 2 and TBX 3 concepts, languages, synonyms, definitions, context, notes, status, and descriptive fields |
| MultiTerm XML | .xml | Trados MultiTerm concept identifiers, language groups, term groups, definitions, context, term notes, and references |
| Excel workbook | .xlsx | Source and target terms plus optional definition, context, and notes columns |
| Delimited text | .csv, .tsv | The same column mapping as Excel |
Lectio preserves alternate target terms as separate glossary senses. Imported terms are confirmed by default. Turn off Import as confirmed terms in the preview when the terminology still needs review.
Multilingual Files#
Lectio detects every language in the source file and asks you to choose a primary source language. It then creates one translation memory or glossary for each selected target language.
Language codes use Lectio's canonical BCP 47 language list. Common variants such as en_US and zh_Hans are normalized automatically. The migration workspace asks you to map unknown or ambiguous codes.
Spreadsheet Structure#
Spreadsheets and delimited files need a header row. Smartcat spreadsheets commonly place the source language in the first column and target languages in later columns.
During mapping, choose:
- The resource type.
- The source language and source column.
- One or more target languages and their columns.
- An optional context column.
Lectio never evaluates spreadsheet formulas. It reads the stored result as text.
Unsupported Native Databases#
Lectio does not directly read these proprietary database or project formats:
- Trados
.sdltmtranslation-memory databases. - Trados
.sdltbterminology databases. - CAT project packages.
- Bilingual XLIFF used as a project working file.
Export .sdltm as TMX. Export .sdltb through MultiTerm as MultiTerm XML, TBX, or Excel. This produces a portable file that Lectio can inspect before import.
Encoding and XML Safety#
Lectio detects UTF-8, UTF-16 little-endian, UTF-16 big-endian, and Windows-1252 text. You can correct the detected encoding during mapping. UTF-8 remains the best choice for portable multilingual CSV and TSV files.
The importer rejects XML document type declarations, external entities, unsafe nesting, oversized metadata, cells or segments, and files that exceed processing safety limits. These checks protect your workspace without changing valid CAT exports.
Translation Memory Exports#
Lectio exports translation memories as TMX. The export includes source text, target text, and supported entry metadata.
Glossary Exports#
Lectio exports glossaries as CSV with source term, target term, context, and confirmation state.
For the complete migration workflow, see Migrate Resources From Another CAT Tool.