add ods support 0.22 release

Author: kindly

Cargo.lock

@@ -1,6 +1,6 @@
 [package]
 name = "flatterer"
-version = "0.21.1"
+version = "0.22.0"
 authors = ["David Raznick <[email protected]>"]
 edition = "2021"
 license = "MIT"
@@ -14,9 +14,9 @@ serde_json = { version = "1.0.83", features = ["preserve_order"] }
 pyo3 = { version = "0.18.3", features = ["extension-module", "eyre"] }
 eyre = "0.6.8"
 #libflatterer={path = "../libflatterer"}
-libflatterer = "0.21.1"
+libflatterer = "0.22.0"
 
-flatterer-web = "0.21.2"
+flatterer-web = "0.22.0"
 #flatterer-web={path = "../flatterer-web"}
 
 env_logger = "0.10.1"

Cargo.toml

@@ -4,7 +4,13 @@ All notable changes to this project will be documented in this file.
 
 and this project adheres to [Semantic Versioning](http://semver.org/).
 
-## [0.21.1] - 2024-04-5
+## [0.22.0] - 2024-04-10
+
+### New
+- ODS output
+
+
+## [0.21.1] - 2024-04-05
 
 ### Changed
 - UI backend changed to actix-web

docs/changelog.md

@@ -18,11 +18,11 @@
 # -- Project information -----------------------------------------------------
 
 project = 'flatterer'
-copyright = '2021-2022, David Raznick'
+copyright = '2021-2025, David Raznick'
 author = 'David Raznick'
 
 # The full version, including alpha/beta/rc tags
-release = '0.19'
+release = '0.22.0'
 
 myst_heading_anchors = 3
 

docs/conf.py

@@ -15,6 +15,7 @@ Options:
   --web                       Load web based version
   --csv / --nocsv             Output CSV files, default true
   --xlsx / --noxlsx           Output XLSX file, default false
+  --ods / --noods             Output ODS file, default false
   --sqlite / --nosqlite       Output sqlite.db file, default false
   --parquet / --noparquet     Output directory of parquet files, default false
   --postgres TEXT             Connection string to postgres. If supplied will
@@ -74,6 +75,8 @@ Options:
 
 **XLSX:**  Output xlsx file to `<OUTPUT_DIRECTORY>/output.xlsx`.
 
+**ODS:**  Output xlsx file to `<OUTPUT_DIRECTORY>/output.ods`.
+
 **SQLITE:**  Output sqlite file to  `<OUTPUT_DIRECTORY>/sqlite.db`.
 
 **PARQUET:**  Output parquet files in `<OUTPUT_DIRECTORY>/parquet/`.
@@ -92,6 +95,11 @@ flatterer --nocsv INPUT_FILE OUTPUT_DIRECTORY
 flatterer --xlsx INPUT_FILE OUTPUT_DIRECTORY
 ```
 
+**ods output:**
+```bash 
+flatterer --ods INPUT_FILE OUTPUT_DIRECTORY
+```
+
 **sqlite output:**
 ```bash 
 flatterer --sqlite INPUT_FILE OUTPUT_DIRECTORY
@@ -129,7 +137,7 @@ flatterer.flatten('inputfile.json', 'ouput_dir', csv=False, sqlite=True, xlsx=Tr
 
 Name of the table that represents data at the root of the JSON object.  
 
-For CSV will create `<OUTPUT_DIRECTORY>/csv/<main_table_name>.csv` and for XLSX will be the first tab name.
+For CSV will create `<OUTPUT_DIRECTORY>/csv/<main_table_name>.csv` and for XLSX/ODS will be the first tab name.
 
 For CLI defaults to name of input file without the file ending and for python defaults to `main`.
 
@@ -567,8 +575,8 @@ flatterer.flatten('inputfile.jl', 'ouput_dir', fields='fields.csv', only_fields=
 
 Path to tables CSV file.  The file can be used for:
 
-* Changing the sheet order in xlsx output. 
-* Giving the tables (and xlsx sheets) a new name by using `table_title`
+* Changing the sheet order in xlsx/ods output. 
+* Giving the tables (and xlsx/ods sheets) a new name by using `table_title`
 * Removing unwanted tables when using the [](#only-tables) option.
 
 The CSV file needs the following headers: 

docs/options.md

@@ -1,9 +1,9 @@
 # Outputs
 
-All of CSV, XLSX, PARQUET and SQLITE data can be produced with the following.
+All of CSV, XLSX, ODS, PARQUET and SQLITE data can be produced with the following.
 
 ```bash
-flatterer games.json games_dir --xlsx --csv --sqlite --parquet
+flatterer games.json games_dir --xlsx --ods --csv --sqlite --parquet
 ```
 
 This will create the directory structure:
@@ -21,6 +21,7 @@ games_dir/
 ├── tables.csv
 ├── sqlite.db
 ├── output.xlsx
+├── output.ods
 ├── postgresql
 │   ├── postgresql_load.sql
 │   └── postgresql_schema.sql
@@ -37,6 +38,10 @@ The `csv` directory contains a CSV file representing a relational table.  A new
 
 The XLSX output can be found in the output.xlsx file. It contains the same data as the CSV files with a sheet per table.  Sheet names are truncated to not go over the allowed limit.
 
+## ODS
+
+The ODS output can be found in the output.ods file. File is loaded into memory, this could cause issues for large files. Sheet names are truncated to not go over the allowed limit.
+
 ## SQLITE
 
 A `sqlite.db` is created in the output directory where each csv file is put in its own table.

docs/outputs.md

@@ -60,6 +60,7 @@ def flatten(
     output_dir='',
     csv=True,
     xlsx=False,
+    ods=False,
     sqlite=False,
     parquet=False,
     dataframe=False,
@@ -139,7 +140,7 @@ def flatten(
         s3 = True if output_dir.startswith("s3://") else False
 
         if method == 'flatten':
-            flatten_rs(input, output_dir, csv, xlsx, sqlite, parquet,
+            flatten_rs(input, output_dir, csv, xlsx, ods, sqlite, parquet,
                        main_table_name, tables_csv, only_tables, fields_csv, only_fields,
                        inline_one_to_one, path_separator, preview, 
                        table_prefix, id_prefix, emit_obj, force,  
@@ -153,7 +154,7 @@ def flatten(
             if s3:
                 raise AttributeError("s3 output not available when supplying an iterator")
 
-            iterator_flatten_rs(bytes_generator(input), output_dir, csv, xlsx, sqlite, parquet,
+            iterator_flatten_rs(bytes_generator(input), output_dir, csv, xlsx, ods, sqlite, parquet,
                        main_table_name, tables_csv, only_tables, fields_csv, only_fields,
                        inline_one_to_one, path_separator, preview, 
                        table_prefix, id_prefix, emit_obj, force,  
@@ -191,6 +192,9 @@ def flatten(
         if xlsx:
             output['xlsx'] = os.path.join(output_dir, 'output.xlsx')
 
+        if ods:
+            output['ods'] = os.path.join(output_dir, 'output.ods')
+
         return output
 
     finally:
@@ -207,6 +211,7 @@ def iterator_flatten(*args, **kw):
 @click.option('--web', default=False, is_flag=True ,help='Load web based version')
 @click.option('--csv/--nocsv', default=True, help='Output CSV files, default true')
 @click.option('--xlsx/--noxlsx', default=False, help='Output XLSX file, default false')
+@click.option('--ods/--noods', default=False, help='Output ODS file, default false')
 @click.option('--sqlite/--nosqlite', default=False, help='Output sqlite.db file, default false')
 @click.option('--parquet/--noparquet', default=False, help='Output directory of parquet files, default false')
 @click.option('--postgres', default="", help='Connection string to postgres. If supplied will load data into postgres')
@@ -258,6 +263,7 @@ def cli(
     web=False,
     csv=True,
     xlsx=False,
+    ods=False,
     sqlite=False,
     sqlite_path='',
     parquet=False,
@@ -328,6 +334,7 @@ def cli(
                 output_directory,
                 csv=csv,
                 xlsx=xlsx,
+                ods=ods,
                 sqlite=sqlite,
                 parquet=parquet,
                 postgres=postgres,

flatterer/__init__.py

@@ -43,6 +43,7 @@ fn flatterer(_py: Python, m: &PyModule) -> PyResult<()> {
         output_dir: String,
         csv: bool,
         xlsx: bool,
+        ods: bool,
         sqlite: bool,
         parquet: bool,
         main_table_name: String,
@@ -85,6 +86,7 @@ fn flatterer(_py: Python, m: &PyModule) -> PyResult<()> {
 
         op.csv = csv;
         op.xlsx = xlsx;
+        op.ods = ods;
         op.sqlite = sqlite;
         op.parquet = parquet;
         op.main_table_name = main_table_name;
@@ -140,6 +142,7 @@ fn flatterer(_py: Python, m: &PyModule) -> PyResult<()> {
         output_dir: String,
         csv: bool,
         xlsx: bool,
+        ods: bool,
         sqlite: bool,
         parquet: bool,
         main_table_name: String,
@@ -178,6 +181,7 @@ fn flatterer(_py: Python, m: &PyModule) -> PyResult<()> {
 
         options.csv = csv;
         options.xlsx = xlsx;
+        options.ods = ods;
         options.sqlite = sqlite;
         options.parquet = parquet;
         options.main_table_name = main_table_name;
@@ -223,6 +227,10 @@ fn flatterer(_py: Python, m: &PyModule) -> PyResult<()> {
                 log::warn!("XLSX output not supported in multi threaded mode");
                 options.xlsx = false;
             }
+            if options.ods {
+                log::warn!("ODS output not supported in multi threaded mode");
+                options.ods = false;
+            }
             if final_output_path.is_dir() {
                 if options.force {
                     remove_dir_all(&final_output_path)?;