Blogging

As I switched from IBM Connection to WordPress for my blog, I started thinking about my existing content. Was there a way to move them all over without having to manually copy and paste and recreate all 268 entries?

Well, there is, and this is how I did it, using just a  few tools. First I used Wget to retrieve my old blog. This put all the posts on one folder (entries), and all images in another (resource). It was then a simple task to write a Lotusscript agent that processed each file in that folder and read the content, parsed out the title, date originally posted and HTML for the blog post itself. I put that data into separate Notes documents, after performing some cleanup and string replacement.

I had already moved all images to a filer on my primary web server, so I performed a replace of the image URLs in the HTML, to have any images pointing to their new location. I also had to fix some special characters and replace them with the corresponding HTML entities.

Now when I had all the data, I just wrote another agent to export the data out again, to create a CSV file. I then installed a CSV importer in my WordPress blog and used to to import the file I just created.

After a few tweaks I performed a successful import. Later I realized I had missed a few special characters, so I had to fix those entries, but we are talking about 4 or 5, out of 268 entries.

If there is an interest, I might clean up the code a little and create a nicer UI (right now many of the values like path and URL are hard-coded) and then release the code if anyone else is planning to go through the same exercise. Below is the existing code to read the blog entries into a simple Notes database.

Option Public
Option Declare

Dim entrydir As String
Dim resourcedir As String

Sub Initialize
	Dim filename As String
	Dim cnt List As Integer
	Dim blogentry List As String
	Dim tst As Variant 

	entrydir = "D:\BleedYellowBlog\www.bleedyellow.com\blogs\texasswede\entry\"
	resourcedir = "D:\BleedYellowBlog\www.bleedyellow.com\blogs\texasswede\resource\"

	cnt("Total") = 0
	filename = Dir$(entrydir + "*.*")
	Do While fileName <> ""
		blogentry(filename) = entrydir + filename
		cnt("Total") = cnt("Total") + 1
		fileName = Dir$()
	Loop

	cnt("Processed") = 0 
	ForAll be In blogentry 
		cnt("Processed") = cnt("Processed") + 1
		Print "Processing " & cnt("Processed") & " of " & cnt("Total")  
		Call ProcessBlogEntry(ListTag(be),be)	
	End ForAll
End Sub

Function FixHTML(html As String) As String
	Dim tmp As String

	tmp = Replace(html,_
"https://www.bleedyellow.com/blogs/texasswede/resource/",_
"http://www.texasswede.com/blogfiles/resource/")
	tmp = Replace(tmp,_
"http://www.bleedyellow.com/blogs/texasswede/resource/",_
"http://www.texasswede.com/blogfiles/resource/")
	tmp = Replace(tmp,"/BLOGS_UPLOADED_IMAGES/","/uploaded_images/")
	tmp = Replace(tmp,"´",|"&acute;"|)
	tmp = Replace(tmp,"’","&acute;")
	tmp = Replace(tmp,"“",|&quot;|)
	tmp = Replace(tmp,"”",|&quot;|)
	tmp = Replace(tmp,"…",|"..."|)
	tmp = Replace(tmp,"<wbr>",||)
	tmp = Replace(tmp,"> < ",|>&anp;nbsp;< |) 	
        FixHTML = tmp 
End Function 

Function ProcessBlogEntry(filename As String, localpath As String) As Boolean 	
        Dim session As New NotesSession 
	Dim db As NotesDatabase
        Dim blogentry As NotesDocument 	
        Dim rtitem As NotesRichTextItem
        Dim siteurl As String  	
        Dim html List As String
        Dim tmp As String
        Dim import As Boolean
        Dim titlesection As Boolean
        Dim row As Integer
        Dim currow As Integer  	
        Dim titletext As string
        Dim htmltext As String
        Dim title As String
        Dim posteddate As String
        import = False 	
        titlesection = False
        row = 0 	
        Open localpath For Input As #1 charset="UTF-8"
        Do Until EOF(1)
            Line Input #1, tmp
            If InStr(tmp,|class="entryContentContainer"|) > 0 Then
	 	import = True		
	    End If
	    If import = True Then
		If InStr(LCase(tmp),|<!-- rating -->|) > 0 Then
			import = False		
		End If
 	    End If
	    If InStr(LCase(tmp),|<!-- entry title and info -->|) > 0 Then
		titlesection = True		
	    End If
	    If titlesection = True Then
		If InStr(LCase(tmp),|<!-- user name, date, meta info -->|) > 0 Then
			titlesection = False
		End If
	    End If
	    If titlesection = True Then
		titletext = titletext + tmp
	    End If
	    If InStr(LCase(tmp),|blogsdate.date.localize|) > 0 Then
		posteddate = StrLeft(StrRight(tmp,"localize ("),"));")
	    End If
	    If import = True Then
		row = row + 1
	 	html(CStr(row)) = tmp
	    End If
	Loop
	Close #1

	Set db = session.CurrentDatabase 
	Set blogentry = New NotesDocument(db)
	blogentry.Form = "Blog Entry"
	title = Replace(FullTrim(StrLeft(StrRight(titletext,"<h4>"),"</h4>")),"@amp;quot;",|"|)
	Set rtitem = New NotesRichTextItem(blogentry,"Content") 
	posteddate = Format$(JSMillisecondsToLSDate(CDbl(posteddate)),"mm/dd/yyyy hh:nn") + " GMT"
	siteurl = "http://www.bleedyellow.com/blogs/texasswede/"

	Call blogentry.ReplaceItemValue("Title", title)
	Call blogentry.ReplaceItemvalue("PostedDate", posteddate)
	Call blogentry.ReplaceItemValue("OriginalURL", siteurl + filename)
	currow = 0
	ForAll t In html
		currow = currow + 1
		If InStr(t,	|class="entryContentContainer"|)>0 Then
			' Do nothing				
		Else
			If currow < row-2 Then
				Call rtitem.AppendText(fixhtml(t))
				Call rtitem.AddNewLine(1,true)
			End If
		End If
	End ForAll
	Call blogentry.ComputeWithForm(True,False)
	Call blogentry.Save(True,True)

End Function

Function JSMillisecondsToLSDate(millis As Double) As Variant
	Dim ndt As NotesDateTime
	Dim zoneOffset As Integer
	Dim jsEpochDouble As Double, adjustedEpochDouble As Double, millisDateDouble As Double

	%REM
	JavaScript millisecond values are based on GMT
	but writable LotusScript date/time values are local.
	We need to know the local timezone offset from GMT,
	and for that we need a NotesDateTime object
	with both date and time components
	%END REM

	Set ndt = New NotesDateTime(Now)
	zoneOffset = ndt.TimeZone

	'The JavaScript epoch is midnight (day start) January 1, 1970 GMT
	jsEpochDouble = CDbl(DateNumber(1970,1,1))

	'Adjust epoch to local time
	adjustedEpochDouble = jsEpochDouble - (zoneOffset/24)

	'There are 86400000 milliseconds in a day
	millisDateDouble = adjustedEpochDouble + (millis / 86400000)
	JSMillisecondsToLSDate = CDat(millisDateDouble)
End Function

And here is the  agent to export the documents to a CSV file that can be imported into a WordPress blog using the CSV import plugin.

Option Public
Option Declare

Sub Initialize
	Dim session As New NotesSession
	Dim db As NotesDatabase
	Dim view As NotesView
	Dim doc As NotesDocument
	Dim filename As String

	filename = "d:\bleedyellow.csv"
	Open filename For Output As #1
	Print #1, |"csv_post_title","csv_post_post",| + _ 
                  |"csv_post_type","csv_post_excerpt",| + _ 
                  |"csv_post_categories","csv_post_tags",| + _ 
                  |"csv_post_date","custom_field_1","custom_field_2"|
	Set db = session.Currentdatabase
	Set view = db.GetView("By Title")
	Set doc = view.GetFirstDocument
	Do Until doc Is Nothing
		Print #1, GetCSV(doc)
		Set doc = view.GetNextDocument(doc)	
	Loop
	Close #1
End Sub

Function GetCSV(doc As NotesDocument) As String
	Dim rtitem As NotesRichTextItem 
	Dim tmp As String
	Dim content As String

	Set rtitem = doc.Getfirstitem("Content")
	content = Replace(FullTrim(rtitem.GetUnformattedText()),|"|,|""|)
	tmp = |"| + Replace(doc.GetItemValue("Title")(0),|"|,|""|) + |",|
	tmp = tmp + |"| + content + |",|
	tmp = tmp + ",,"
	tmp = tmp +|"| + "Old Blog Post" + |",|
	tmp = tmp +|"| + doc.GetItemValue("Tags")(0) + |",|
	tmp = tmp +|"| + doc.GetItemValue("PostedDate")(0) + |",,,|

	GetCSV = tmp
End Function