ssdiff

Show the difference between two tables/databases/spreadsheets.

Usage

• ssdiff [options] FILE1 FILE2

Index

• Option summary
• Option details
• Examples
• Patch formats
• Database/spreadsheet file formats
• Version

Option summary

• --act=ACT
• --apply
• --bid=COLUMN
• --context=N
• --default-table=TABLE
• --exclude-column=COLUMN
• --fixed-columns
• --format=FORMAT
• --git
• --head-trimmed
• --headerless
• --help
• --id=COLUMN
• --include-column=COLUMN
• --input-formats
• --low-memory
• --named
• --omit-format-name
• --omit-sheet-name
• --output=OUTPUTFILE
• --parent=PARENT
• --patch-formats
• --scan-for-patch
• --table=TABLE
• --tail-trimmed
• --unordered
• --variant=VARIANT

Option details

--act=ACT
filter for an action of a particular type (update, insert, delete, none, schema)

--apply
apply difference between FILE1 and FILE2 immediately to FILE1

--bid=COLUMN
boost a column (repeat option for multiple columns)

--context=N
Number of rows of context before and after changes for highlighter diffs ("all" to include all rows)

--default-table=TABLE
name to use when a table name is needed and not supplied

--exclude-column=COLUMN
exclude the specified column even if changed

--fixed-columns
ignore new or removed columns

--format=FORMAT
set difference format for output

--git
expect git-compatible parameters (path old-file old-hex old-mode new-file new-hex new-mode)

--head-trimmed
ignore rows removed at the beginning of a table (such as a log file)

--headerless
treat any embedded column names as regular parts of the table (for formats like CSV)

--help
show how to use this program

--id=COLUMN
set primary key (repeat option for multi-column key)

--include-column=COLUMN
include the specified column even if unchanged

--input-formats
list supported input database formats

--low-memory
prioritize low memory usage over speed

--named
trust names of columns, omitting checks for column renames

--omit-format-name
omit any version-dependent header from diff

--omit-sheet-name
omit any sheet/table name from diff

--output=OUTPUTFILE
direct output to this file (default is standard output)

--parent=PARENT
use named workbook/database as common ancestor in difference calculations

--patch-formats
list supported patch formats

--scan-for-patch
check if FILE2 looks like a patch, and if so, apply it

--table=TABLE
filter for a named table of a workbook/database (repeat option for multiple tables)

--tail-trimmed
ignore rows removed at the end of a table (such as a log file)

--unordered
treat order of rows as unimportant

--variant=VARIANT
set the desired dialect when using a poorly defined output format (currently for SQL, available variants are: sqlite, access)

Examples

You can generate test file(s) for the examples that follow:

ssdiff --test-file numbers.csv
ssdiff --test-file numbers.sqlite
ssdiff --test-file numbers_buggy.csv
ssdiff --test-file numbers_buggy.sqlite

Example 1

ssdiff numbers_buggy.csv numbers.csv

Compare two tables. Output goes to standard output.

Example 2

ssdiff --unordered numbers_buggy.csv numbers.csv

Compare two tables, neglecting row order.

Example 3

ssdiff --format sql numbers_buggy.sqlite numbers.sqlite

Compare two databases, with output in SQL format.

Example 4

ssdiff --format hilite --output review.csv numbers_buggy.csv numbers.csv

Generate tabular diff for eyeballing. If ssdiff is compiled with gnumeric support, and output format is *.xls, color highlighting is added.

Patch formats

• tdiff: [default] reminiscent of the standard unix diff format for text
• csv: csv-compatible diff format
• hilite: colorful spreadsheet format
• index: tabular output showing relationship between rows and columns
• novel: mark all shared rows - remaining rows are unmatched
• ops: summarize modified rows in a table
• raw: verbose diff format for debugging
• review: spreadsheet diff format suitable for quickly accepting or rejecting changes
• sql: SQL format (data diffs only)
• stats: produce statistics on table changes

Database/spreadsheet file formats

CSV: plain-text delimiter-separated family of formats

• .csv: Comma-separated values
• .tsv: Tab-separated values
• .ssv: Semicolon-separated values
• .json: {
  "type": "csv", // CSV family
  "file": "fname.dsv", // File name
  "delimiter": "|" // Delimiter character
  }

SQLITE: file-based database

• .sqlite: Sqlite database file
• .json: {
  "type": "sqlite", // Sqlite family
  "file": "fname.db" // File name
  }
• dbi:sqlite:fname.db (Force sqlite interpretation)

SQLITEXT: sqlite-format sql dump

• .sqlitext: SQL dump of Sqlite database
• .json: {
  "type": "sqlitext", // Sqlitext family
  "file": "fname.sql" // File name
  }
• dbi:sqlitext:fname.sql (Force sqlitext interpretation)

JSONBOOK: Spreadsheet formats in json

• .jsonbook: Json spreadsheet
• .json: {
  "type": "jsonbook", // Json spreadsheet
  "file": "fname.sheet", // File name
  "hash": "1" // Generate hash/dict/objects for each row when schema available
  }
• dbi:jsonbook:fname.sheet (Force Json spreadsheet interpretation)

GNUMERIC: Spreadsheet formats (via gnumeric)

• .xls: Excel spreadsheet
• .xlsx: Excel spreadsheet
• .gnumeric: Gnumeric spreadsheet
• .ods: OpenOffice/LibreOffice spreadsheet
• .json: {
  "type": "gnumeric", // Gnumeric family
  "file": "fname.sheet" // File name
  }
• dbi:gnumeric:fname.sheet (Force Gnumeric interpretation)

MDB: Access database format (via Mdbtools, READ-ONLY)

• .mdb: Access database file
• .json: {
  "type": "access", // Access family
  "file": "fname.db" // File name
  }
• dbi:access:fname.db (Force Access interpretation)

JMDB: Access database format (via Jackcess)

• .mdb: Access database file
• .json: {
  "type": "access", // Jackcess/access family
  "file": "fname.db" // File name
  }
• dbi:jackess:fname.db (Force Access interpretation)

MYSQL: database connector

• .json: {
  "type": "mysql", // MYSQL connector
  "database": "db_name", // Database name
  "host": "localhost", // Host name
  "port": "1111", // Port number
  "username": "root", // Username
  "password": "****" // Password
  }
• dbi:mysql:database_name (Use default port, username, etc)
• dbi:mysql:database_name:username=USER:password=PASS
• dbi:mysql:database_name:host=HOST:port=PORT

SOCIALCALC: SocialCalc format (via mozjs)

• .socialcalc: SocialCalc spreadsheet file
• .sc: SocialCalc spreadsheet file
• .json: {
  "type": "socialcalc", // SocialCalc family
  "file": "sheet.txt" // File name
  }
• dbi:socialcalc:sheet.txt (Force SocialCalc interpretation)

Version

ssdiff version 0.6.5