update dependencies + small fixes

lu4p · Mar 21, 2020 · dcbc6f5 · dcbc6f5
1 parent 583a2e1
commit dcbc6f5
Show file tree

Hide file tree

Showing 9 changed files with 60 additions and 36 deletions.
diff --git a/cat.go b/cat.go
@@ -23,8 +23,8 @@ func File(filename string) (string, error) {
 
 // FromBytes converts a []bytes representation of a document to text
 func FromBytes(data []byte) (string, error) {
-	mime, _ := mimetype.Detect(data)
-	switch mime {
+	mime := mimetype.Detect(data)
+	switch mime.String() {
 	case "application/vnd.oasis.opendocument.text":
 		return odtxt.BytesToStr(data)
 	case "application/pdf":

diff --git a/docxtxt/docxreader.go b/docxtxt/docxreader.go
@@ -9,7 +9,7 @@ import (
 	"regexp"
 )
 
-//docx zip struct
+// docx zip struct
 type docx struct {
 	zipFileReader *zip.ReadCloser
 	Files         []*zip.File
@@ -28,7 +28,6 @@ func ToStr(filename string) (string, error) {
 		return "", err
 	}
 	return BytesToStr(content)
-
 }
 
 // BytesToStr converts a []byte representation of .docx document file to string
@@ -97,7 +96,7 @@ func (d *docx) GenWordsList() {
 	d.listP(xmlData)
 }
 
-//get w:t value
+// get w:t value
 func (d *docx) getT(item string) {
 	var subStr string
 	data := item
@@ -131,7 +130,7 @@ func hasP(data string) bool {
 func (d *docx) listP(data string) {
 	var result []string
 	re := regexp.MustCompile(`(?U)<w:p>(.*)</w:p>`)
-	for _, match := range re.FindAllStringSubmatch(string(data), -1) {
+	for _, match := range re.FindAllStringSubmatch(data, -1) {
 		result = append(result, match[1])
 	}
 	for _, item := range result {

diff --git a/go.mod b/go.mod
@@ -1,9 +1,10 @@
 module github.com/lu4p/cat
 
-go 1.13
+go 1.14
 
 require (
 	github.com/EndFirstCorp/peekingReader v0.0.0-20171012052444-257fb6f1a1a6
-	github.com/gabriel-vasile/mimetype v0.3.21
-	github.com/lu4p/unipdf/v3 v3.3.2
+	github.com/gabriel-vasile/mimetype v1.0.4
+	github.com/lu4p/unipdf/v3 v3.5.1-0.20200321134028-22d47cef6c06
+	golang.org/x/image v0.0.0-20200119044424-58c23975cae1 // indirect
 )
diff --git a/go.sum b/go.sum
@@ -1,30 +1,35 @@
 github.com/EndFirstCorp/peekingReader v0.0.0-20171012052444-257fb6f1a1a6 h1:t27CGFMv8DwGwqRPEa2VNof5I/aZwO6q2gfJhN8q0U4=
 github.com/EndFirstCorp/peekingReader v0.0.0-20171012052444-257fb6f1a1a6/go.mod h1:zpqkXxDsVfEIUZEWvT9yAo8OmRvSlRrcYQ3Zs8sSubA=
+github.com/adrg/strutil v0.1.0/go.mod h1:pXRr2+IyX5AEPAF5icj/EeTaiflPSD2hvGjnguilZgE=
+github.com/adrg/sysfont v0.1.0/go.mod h1:DzISco90USPZJ+lmtpuz1SOTn1fih6YyB0KG2TEP/0U=
+github.com/adrg/xdg v0.2.1/go.mod h1:ZuOshBmzV4Ta+s23hdfFZnBsdzmoR3US0d7ErpqSbTQ=
+github.com/boombuler/barcode v1.0.0 h1:s1TvRnXwL2xJRaccrdcBQMZxq6X7DvsMogtmJeHDdrc=
 github.com/boombuler/barcode v1.0.0/go.mod h1:paBWMcWSl3LHKBqUq+rly7CNSldXjb2rDl3JlRe0mD8=
+github.com/davecgh/go-spew v1.1.0 h1:ZDRjVQ15GmhC3fiQ8ni8+OwkZQO4DARzQgrnXU1Liz8=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
-github.com/gabriel-vasile/mimetype v0.3.21 h1:Mc81ydjjIFN3Ir12WJ4myhnMs6cFAIlthU7MKY6XAIk=
-github.com/gabriel-vasile/mimetype v0.3.21/go.mod h1:6CDPel/o/3/s4+bp6kIbsWATq8pmgOisOPG40CJa6To=
+github.com/gabriel-vasile/mimetype v1.0.4 h1:uBejfH8l3/2f+5vjl1e4xIaSyNEhRBZ5N/ij7ohpNd8=
+github.com/gabriel-vasile/mimetype v1.0.4/go.mod h1:6CDPel/o/3/s4+bp6kIbsWATq8pmgOisOPG40CJa6To=
+github.com/golang/freetype v0.0.0-20170609003504-e2365dfdc4a0/go.mod h1:E/TSTwGwJL78qG/PmXZO1EjYhfJinVAhrmmHX6Z8B9k=
+github.com/gunnsth/pkcs7 v0.0.0-20181213175627-3cffc6fbfe83 h1:saj5dTV7eQ1wFg/gVZr1SfbkOmg8CYO9R8frHgQiyR4=
 github.com/gunnsth/pkcs7 v0.0.0-20181213175627-3cffc6fbfe83/go.mod h1:xaGEIRenAiJcGgd9p62zbiP4993KaV3PdjczwGnP50I=
-github.com/lu4p/unipdf/v3 v3.3.2 h1:MI1CmbDjxA2LhAJwfKMH9USo16B0iLU4XakQ/9ZbdI8=
-github.com/lu4p/unipdf/v3 v3.3.2/go.mod h1:BN0DPOet7qvp6n+JbZVUR1MgWNROJMlg8l43dCFRU1A=
+github.com/lu4p/unipdf/v3 v3.5.1-0.20200321134028-22d47cef6c06 h1:r49J8T7xpn6yTAncprN10Ye2dWqJ490TkNBGtRtwZKw=
+github.com/lu4p/unipdf/v3 v3.5.1-0.20200321134028-22d47cef6c06/go.mod h1:748EuXjaKAupkXEtU/wJQX7efgs3/mPvNtrV1kAE6vY=
+github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/testify v1.3.0 h1:TivCn/peBQ7UY8ooIcPgZFpTNSz0Q2U6UrFlUfqbe0Q=
 github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
 golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
+golang.org/x/crypto v0.0.0-20190605123033-f99c8df09eb5 h1:58fnuSXlxZmFdJyvtTFVmVhcMLU6v5fEb/ok4wyqtNU=
 golang.org/x/crypto v0.0.0-20190605123033-f99c8df09eb5/go.mod h1:yigFU9vqHzYiE8UmvKecakEJjdnWj3jj499lnFckfCI=
 golang.org/x/image v0.0.0-20181116024801-cd38e8056d9b h1:VHyIDlv3XkfCa5/a81uzaoDkHH4rr81Z62g+xlnO8uM=
 golang.org/x/image v0.0.0-20181116024801-cd38e8056d9b/go.mod h1:ux5Hcp/YLpHSI86hEcLt0YII63i6oz57MZXIpbrjZUs=
-golang.org/x/lint v0.0.0-20190409202823-959b441ac422/go.mod h1:6SW0HCj/g11FgYtHlgUYUwCkIfeOF89ocIRzGO/8vkc=
-golang.org/x/net v0.0.0-20190311183353-d8887717615a/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
+golang.org/x/image v0.0.0-20200119044424-58c23975cae1 h1:5h3ngYt7+vXCDZCup/HkCQgW5XwmSvR/nA2JmJ0RErg=
+golang.org/x/image v0.0.0-20200119044424-58c23975cae1/go.mod h1:FeLwcggjj3mMvU+oOTbSwawSJRM1uh48EjtB4UJZlP0=
 golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
-golang.org/x/net v0.0.0-20190606173856-1492cefac77f/go.mod h1:HSz+uSET+XFnRR8LxR5pz3Of3rY3CfYBVs4xY44aLks=
-golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
 golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
 golang.org/x/sys v0.0.0-20190412213103-97732733099d/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
-golang.org/x/sys v0.0.0-20190606203320-7fc4e5ec1444/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
 golang.org/x/text v0.3.2 h1:tW2bmiBqwgJj/UpqtC8EpXEZVYOwU0yG4iWbprSVAcs=
 golang.org/x/text v0.3.2/go.mod h1:bEr9sfX3Q8Zfm5fL9x+3itogRgK3+ptLWKqgva+5dAk=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
-golang.org/x/tools v0.0.0-20190311212946-11955173bddd/go.mod h1:LCzVGOaR6xXOjkQ3onu1FJEFr0SW1gC7cKk1uF8kGRs=
-golang.org/x/tools v0.0.0-20190606174628-0139d5756a7d/go.mod h1:/rFqwRUd4F7ZHNgwSSTFct+R/Kf4OFW1sUzUTQQTgfc=
diff --git a/odtxt/odtreader.go b/odtxt/odtreader.go
@@ -9,7 +9,7 @@ import (
 	"io/ioutil"
 )
 
-//Odt zip struct
+// odt zip struct
 type odt struct {
 	zipFileReader *zip.ReadCloser
 	Files         []*zip.File
@@ -46,7 +46,7 @@ func BytesToStr(data []byte) (string, error) {
 	}
 	content, err := d.GetTxt()
 	if err != nil {
-		return "", errors.New("Could not Get Content")
+		return "", errors.New("could not Get Content")
 	}
 	return content, nil
 }
@@ -72,7 +72,7 @@ func openReader(bytesReader *bytes.Reader) (*odt, error) {
 	return &odtDoc, nil
 }
 
-//Read all files contents
+// retrieveFileContents Read all files contents
 func (d *odt) retrieveFileContents(filename string) ([]byte, error) {
 	var file *zip.File
 	for _, f := range d.Files {
@@ -96,7 +96,6 @@ func (d *odt) retrieveFileContents(filename string) ([]byte, error) {
 func (d *odt) GetTxt() (string, error) {
 	xmlData := d.FilesContent["content.xml"]
 	return d.listP(xmlData)
-
 }
 
 // listP for w:p tag value

diff --git a/pdftxt/pdftxt.go b/pdftxt/pdftxt.go
@@ -1,4 +1,4 @@
-//Package pdftxt extracts text from .pdf documents
+// Package pdftxt extracts text from .pdf documents
 package pdftxt
 
 import (
@@ -57,3 +57,4 @@ func BytesToStr(data []byte) (string, error) {
 
 	return out, nil
 }
+
diff --git a/plaintxt/plaintxt.go b/plaintxt/plaintxt.go
@@ -13,7 +13,6 @@ func ToStr(filename string) (string, error) {
 		return "", errors.New("Error while reading file: " + err.Error())
 	}
 	return BytesToStr(content)
-
 }
 
 // BytesToStr converts a []byte representation of a plaintext file to string

diff --git a/rtftxt/rtftxt.go b/rtftxt/rtftxt.go
@@ -19,8 +19,8 @@ func ToStr(filename string) (string, error) {
 	content, err := ioutil.ReadFile(filename)
 	if err != nil {
 		return "", err
-
 	}
+
 	return BytesToStr(content)
 }
 
@@ -228,20 +228,38 @@ func readUntilClosingBrace(r peekingReader.Reader) error {
 	return err
 }
 
-func handleParams(control, param string, text *bytes.Buffer) {
+func handleParams(control, param string, text io.StringWriter) {
 	param = strings.TrimPrefix(param, " ")
 	if param == "" {
 		return
 	}
 	switch control {
-
 	case "fldrslt":
 		text.WriteString(param)
-	case "acccircle", "acccomma", "accdot", "accnone", "accunderdot", "animtextN", "b", "caps", "cbN", "cchsN ", "cfN", "charscalexN", "csN", "dnN", "embo", "expndN", "expndtwN ", "fittextN", "fN", "fsN", "i", "kerningN ", "langfeN", "langfenpN", "langN", "langnpN", "ltrch", "noproof", "nosupersub ", "outl", "plain", "rtlch", "scaps", "shad", "strike", "sub ", "super ", "ul", "ulcN", "uld", "uldash", "uldashd", "uldashdd", "uldb", "ulhwave", "ulldash", "ulnone", "ulth", "ulthd", "ulthdash", "ulthdashd", "ulthdashdd", "ulthldash", "ululdbwave", "ulw", "ulwave", "upN", "v", "webhidden":
+	case "acccircle", "acccomma", "accdot", "accnone", "accunderdot",
+		"animtextN", "b", "caps", "cbN", "cchsN ", "cfN", "charscalexN",
+		"csN", "dnN", "embo", "expndN", "expndtwN ", "fittextN", "fN",
+		"fsN", "i", "kerningN ", "langfeN", "langfenpN", "langN", "langnpN",
+		"ltrch", "noproof", "nosupersub ", "outl", "plain", "rtlch", "scaps",
+		"shad", "strike", "sub ", "super ", "ul", "ulcN", "uld", "uldash",
+		"uldashd", "uldashdd", "uldb", "ulhwave", "ulldash", "ulnone", "ulth",
+		"ulthd", "ulthdash", "ulthdashd", "ulthdashdd", "ulthldash", "ululdbwave", "ulw", "ulwave", "upN", "v", "webhidden":
 		text.WriteString(param)
 
 	// Paragraph Formatting Properties
-	case "aspalpha", "aspnum", "collapsed", "contextualspace", "cufiN", "culiN", "curiN", "faauto", "facenter", "fafixed", "fahang", "faroman", "favar", "fiN", "hyphpar ", "indmirror", "intbl", "itapN", "keep", "keepn", "levelN", "liN", "linN", "lisaN", "lisbN", "ltrpar", "nocwrap", "noline", "nooverflow", "nosnaplinegrid", "nowidctlpar ", "nowwrap", "outlinelevelN ", "pagebb", "pard", "prauthN", "prdateN", "qc", "qd", "qj", "qkN", "ql", "qr", "qt", "riN", "rinN", "rtlpar", "saautoN", "saN", "sbautoN", "sbN", "sbys", "slmultN", "slN", "sN", "spv", "subdocumentN ", "tscbandhorzeven", "tscbandhorzodd", "tscbandverteven", "tscbandvertodd", "tscfirstcol", "tscfirstrow", "tsclastcol", "tsclastrow", "tscnecell", "tscnwcell", "tscsecell", "tscswcell", "txbxtwalways", "txbxtwfirst", "txbxtwfirstlast", "txbxtwlast", "txbxtwno", "widctlpar", "ytsN":
+	case "aspalpha", "aspnum", "collapsed", "contextualspace",
+		"cufiN", "culiN", "curiN", "faauto", "facenter",
+		"fafixed", "fahang", "faroman", "favar", "fiN", "hyphpar ",
+		"indmirror", "intbl", "itapN", "keep", "keepn", "levelN", "liN",
+		"linN", "lisaN", "lisbN", "ltrpar", "nocwrap", "noline", "nooverflow",
+		"nosnaplinegrid", "nowidctlpar ", "nowwrap", "outlinelevelN ", "pagebb",
+		"pard", "prauthN", "prdateN", "qc", "qd", "qj", "qkN", "ql", "qr", "qt",
+		"riN", "rinN", "rtlpar", "saautoN", "saN", "sbautoN", "sbN", "sbys",
+		"slmultN", "slN", "sN", "spv", "subdocumentN ", "tscbandhorzeven",
+		"tscbandhorzodd", "tscbandverteven", "tscbandvertodd", "tscfirstcol",
+		"tscfirstrow", "tsclastcol", "tsclastrow", "tscnecell", "tscnwcell",
+		"tscsecell", "tscswcell", "txbxtwalways", "txbxtwfirst", "txbxtwfirstlast",
+		"txbxtwlast", "txbxtwno", "widctlpar", "ytsN":
 		text.WriteString(param)
 
 	// Section Formatting Properties
@@ -253,7 +271,8 @@ func handleParams(control, param string, text *bytes.Buffer) {
 		text.WriteString(param)
 
 	// Special Characters
-	case "-", ":", "_", "{", "|", "}", "~", "bullet", "chatn", "chdate", "chdpa", "chdpl", "chftn", "chftnsep", "chftnsepc", "chpgn", "chtime", "column", "emdash", "emspace ", "endash", "enspace ", "lbrN", "ldblquote", "line", "lquote", "ltrmark", "page", "par", "qmspace", "rdblquote", "row", "rquote", "rtlmark", "sect", "sectnum", "softcol ", "softlheightN ", "softline ", "softpage ", "tab", "zwbo", "zwj", "zwnbo", "zwnj":
+	case "-", ":", "_", "{", "|", "}", "~", "bullet", "chatn", "chdate", "chdpa", "chdpl", "chftn", "chftnsep", "chftnsepc", "chpgn", "chtime", "column", "emdash", "emspace ",
+		"endash", "enspace ", "lbrN", "ldblquote", "line", "lquote", "ltrmark", "page", "par", "qmspace", "rdblquote", "row", "rquote", "rtlmark", "sect", "sectnum", "softcol ", "softlheightN ", "softline ", "softpage ", "tab", "zwbo", "zwj", "zwnbo", "zwnj":
 		text.WriteString(param)
 
 	// Table Definitions

diff --git a/rtftxt/stack_test.go b/rtftxt/stack_test.go
@@ -6,20 +6,21 @@ import (
 
 func TestPushPeekPopLen(t *testing.T) {
 	var s stack
-	s.Push("hello")
-	if s.size != 1 || s.top.value != "hello" || s.top.next != nil {
+	const h = "hello"
+	s.Push(h)
+	if s.size != 1 || s.top.value != h || s.top.next != nil {
 		t.Error("expected valid value", s.size, s.top)
 	}
 
 	if s.Len() != 1 {
 		t.Error("expected correct length")
 	}
 
-	if v := s.Peek(); v != "hello" || s.size != 1 || s.top.value != "hello" || s.top.next != nil {
+	if v := s.Peek(); v != h || s.size != 1 || s.top.value != h || s.top.next != nil {
 		t.Error("expected same value and no size reduction")
 	}
 
-	if v := s.Pop(); v != "hello" {
+	if v := s.Pop(); v != h {
 		t.Error("expected pushed value", v)
 	}