v0.2.3

README.md

@@ -71,6 +71,7 @@ Read    Gzip            |   Yes    |  Yes      |
 Fields ranges           |   Yes    |  Yes      | e.g. `-f 1-4,6`
 **Unselect fileds**     |   Yes    |  --       | e.g. `-1` for excluding first column
 **Fuzzy fields**        |   Yes    |  --       | e.g. `ab*` for columns with prefix "ab"
+Order-specific fields   |   --     |  Yes      | it means `1,2` is different from `2,1`
 Rename columns          |   Yes    |  --       | rename with new name(s) or from existed names
 Sort by multiple keys   |   Yes    |  Yes      | bash sort like operations
 Sort by number          |   Yes    |  --       | e.g. `-k 1:n`
@@ -120,7 +121,8 @@ to be continued...
 
 1. Join multiple files by keys (`join`)
 
-    - `csvtk join -f "username;username;name" names.csv phone.csv adress.csv -k`
+    - All files have same key column: `csvtk join -f id file1.csv file2.csv`
+    - Files have different key columns: `csvtk join -f "username;username;name" names.csv phone.csv adress.csv -k`
 
 ## Contact
 

csvtk/cmd/helper.go

@@ -172,7 +172,7 @@ type Config struct {
 func getConfigs(cmd *cobra.Command) Config {
 	return Config{
 		ChunkSize: getFlagPositiveInt(cmd, "chunk-size"),
-		NumCPUs:   getFlagPositiveInt(cmd, "num-cups"),
+		NumCPUs:   getFlagPositiveInt(cmd, "num-cpus"),
 
 		Delimiter:    getFlagRune(cmd, "delimiter"),
 		OutDelimiter: getFlagRune(cmd, "out-delimiter"),

csvtk/cmd/join.go

@@ -34,8 +34,9 @@ import (
 var joinCmd = &cobra.Command{
 	Use:   "join",
 	Short: "join multiple CSV files by selected fields",
-	Long: ` join multiple CSV files by selected fields.
-Join 2- files to the first one.
+	Long: ` join 2nd and later files to the first file by selected fields.
+	
+Multiple keys supported, but the orders are ignored.
 
 `,
 	Run: func(cmd *cobra.Command, args []string) {
@@ -48,7 +49,13 @@ Join 2- files to the first one.
 		}
 
 		allFields := getFlagSemicolonSeparatedStrings(cmd, "fields")
-		if len(allFields) != len(files) {
+		if len(allFields) == 1 {
+			s := make([]string, len(files))
+			for i := range files {
+				s[i] = allFields[0]
+			}
+			allFields = s
+		} else if len(allFields) != len(files) {
 			checkError(fmt.Errorf("number of fields (%d) should be equal to number of files (%d)", len(allFields), len(files)))
 		}
 		// ignoreCase := getFlagBool(cmd, "ignore-case")
@@ -153,7 +160,8 @@ Join 2- files to the first one.
 
 func init() {
 	RootCmd.AddCommand(joinCmd)
-	joinCmd.Flags().StringP("fields", "f", "", `Semicolon seperated key fields of all files. e.g -f 1,2;2,3 or -f A,B;C,D`)
+	joinCmd.Flags().StringP("fields", "f", "1", "Semicolon seperated key fields of all files, "+
+		"if given one, we think all the files have the same key columns. e.g -f 1;2 or -f A,B;C,D or -f id")
 	joinCmd.Flags().BoolP("ignore-case", "i", false, `ignore case`)
 	joinCmd.Flags().BoolP("fuzzy-fields", "F", false, `using fuzzy fileds, e.g. *name or id123*`)
 	joinCmd.Flags().BoolP("keep-unmatched", "k", false, `keep unmatched data of the first file`)

csvtk/cmd/root.go

@@ -55,7 +55,7 @@ func Execute() {
 
 func init() {
 	RootCmd.PersistentFlags().IntP("chunk-size", "c", 50, `chunk size of CSV reader`)
-	RootCmd.PersistentFlags().IntP("num-cups", "j", runtime.NumCPU(), `number of CPUs to use (default value depends on your computer)`)
+	RootCmd.PersistentFlags().IntP("num-cpus", "j", runtime.NumCPU(), `number of CPUs to use (default value depends on your computer)`)
 
 	RootCmd.PersistentFlags().StringP("delimiter", "d", ",", `delimiting character of the input CSV file`)
 	RootCmd.PersistentFlags().StringP("out-delimiter", "D", ",", `delimiting character of the input CSV file`)

doc/docs/download.md

@@ -8,6 +8,7 @@
 
 - [csvtk v0.2.3](https://github.com/shenwei356/csvtk/releases/tag/v0.2.3)
     - add flag `--colnames` to `cut`
+    - flag `-f` (`--fields`) of `join` supports single value now
 
 ## Installation
 
@@ -38,7 +39,7 @@ You can also add the directory of the executable file to environment variable
 - [csvtk v0.2.1](https://github.com/shenwei356/csvtk/releases/tag/v0.2.1)
     - fix bug of `mutate`
 
-        
+
 <div id="disqus_thread"></div>
 <script>
 /**

doc/docs/usage.md

@@ -7,7 +7,7 @@ Usage
 ```
 Another cross-platform, efficient and practical CSV/TSV toolkit
 
-Version: 0.2.2
+Version: 0.2.3
 
 Author: Wei Shen <shenwei356@gmail.com>
 
@@ -72,7 +72,7 @@ Examples
     7       8       0
     8       1,000   4
 
-    $ csvtk transpose -t digitals.tsv 
+    $ csvtk transpose -t digitals.tsv
     4       1       7       8
     5       2       8       1,000
     6       3       0       4
@@ -169,24 +169,25 @@ Examples
 Usage
 
 ```
-join multiple CSV files by selected fields.
-Join 2- files to the first one.
+join 2nd and later files to the first file by selected fields.
+
+Multiple keys supported, but the orders are ignored.
 
 Usage:
   csvtk join [flags]
 
 Flags:
-  -f, --fields string    Semicolon seperated key fields of all files. e.g -f 1,2;2,3 or -f A,B;C,D
+  -f, --fields string    Semicolon seperated key fields of all files, if given one, we think all the files have the same key columns. e.g -f 1;2 or -f A,B;C,D or -f id (default "1")
   -F, --fuzzy-fields     using fuzzy fileds, e.g. *name or id123*
   -i, --ignore-case      ignore case
   -k, --keep-unmatched   keep unmatched data of the first file
 
-
 ```
 
 Examples:
 
-- `csvtk join -f "username;username;name" names.csv phone.csv adress.csv -k`
+- All files have same key column: `csvtk join -f id file1.csv file2.csv`
+- Files have different key columns: `csvtk join -f "username;username;name" names.csv phone.csv adress.csv -k`
 
 ## rename
 
@@ -303,7 +304,7 @@ Examples
 - Complex sort: `csvtk sort -k region -k age:n -k id:nr`
 
 
-        
+
 <div id="disqus_thread"></div>
 <script>
 /**