Procedure for establishing db of text files.

[JimGatton]

I have a few thousand PLR text files I bought. I would like to load them all up into an existing db or maybe a new db so that I could search for them and have A5 show me in some form the files that would be relevant to my search.

Some of the files, as you can see from the attachments, do contain a line of keywords. Others are just small articles.

A few thousand of these articles loaded manually is going to take a very long time and with no one to pay manually to do this it means that I have to do it.

Is there any way to streamline this approach with A5 along the lines of, "Load all the .txt files in this directory on my harddrive" and have it do so? I played with a similar situation years ago but I think I've forgotten the few things I learned then. I'm not exactly an A5 developer.

I envision searching for my keyword phrase in Alpha, being shown a focused list of articles meeting my criteria and then my manually opening and copying the contents of 5 or 6 files.

Could you give me suggestions on the best approach to handling this? Some creative thinking?

Thanks for all guidance, any and all ideas.

[Stan Mathews]

Re: Procedure for establishing db of text files.

It shouldn't be too hard to develop a script for you that

got a list of the text files in a directory
processed each file in the list
start a new record in a table
------capture the title and put it in a text field
------capture the word count and put it in a numeric field
------capture the summary and put it in a text field
------capture the keywords and put it in a text field
------capture the article body and put it in a memo field
go back an process the next text file


But your first sample doesn't have the same structure as the second where those parts are broken out. How would one process the 1 What is Accounting... file? To write the script described there have to be rules to interpret the contents of the files.

[Stan Mathews]

Re: Procedure for establishing db of text files.

The relevant functions to parse the structured file are....

varf = FILE.TO_STRING("C:\Temp\Acid_Reflux_Diet_Tips.txt")
? "Title:" $ varf
= .T.

? extract_string(varf,"Title:","Word Count:")
=  
Acid Reflux Diet Tips

? extract_string(varf,"Word Count:","Summary:")
= 
266

? extract_string(varf,"Summary:","Keywords:")
= 
There are some foods that seem to encourage heartburn in many people.

? extract_string(varf,"Keywords:","Article Body:")
= 
heartburn, acid, reflux, foods

? words(varf,word_number_get(varf,"Body:",crlf())+1,30000,.F.,crlf())
= No matter what type of acid reflux you suffer from, and no matter the underlying factors to your own particular struggle with heartburn, a change in diet can have remarkable effects upon your management of acid reflux.

There are some foods that seem to encourage heartburn in many people. Among these are general categories such as spicy, high fat and dairy foods. Chocolate, tomatoes and herbal supplements are also known to bring about heartburn like symptoms, particularly if ingested shortly before bedtime.

If you recognize particular foods that seem to effect heartburn, you may find that eating them in the morning instead of closer to bedtime will alleviate your symptoms of heartburn. When eaten during time of wakefulness and activity, the body is better able to contain stomach acids used to digest the food we have eaten.

Many people suffer from heartburn primarily at night time and have found relief by raising the level of the head of their bed. Though simply adding layers of pillows doesn�t have the same effect, raising the entire head end of a bed by at least twelve inches has been shown to lessen heartburn symptoms.

If you still are suffering from acid reflux and diet modification doesn�t seem to help, you should consider seeing a doctor in order to rule out the possibility of structural abnormality or other factors contributing to your acid reflux symptoms.

Even if you end up trying a prescription medicine prescribed by your doctor, it is a very good idea to try and eliminate foods that make heartburn worse, and increase fresh, raw foods.

[JimGatton]

Re: Procedure for establishing db of text files.

If nothing else then the script would have to screen for which of two formats it has encountered and then branch accordingly?

Sorry, I think your second post came through as I was commenting on the first post.

[Stan Mathews]

Re: Procedure for establishing db of text files.

the script would have to screen for which of two formats it has encountered and then branch accordingly

Simple enough. Give me the table name and field names you want to use and the directory in which the files reside.

varf = file.to_string("C:\Temp\1 What Is Accounting Anyway.txt")
? word(varf,1,crlf())
= "What Is Accounting Anyway?"

? alltrim(words(varf,2,30000,.F.,crlf()))
= Anyone who's worked in an office at some point or another has had to go to accounting. They're the people who pay and send out the bills that keep the business running. They do a lot more than that, though. Sometimes referred to as "bean counters" they also keep their eye on profits, costs and losses. Unless you're running your own business and acting as your own accountant, you'd have no way of knowing just how profitable - or not - your business is without some form of accounting.

	No matter what business you're in, even if all you do is balance a checkbook, that's still accounting. It's part of even a kid's life. Saving an allowance, spending it all at once - these are accounting principles. 

	What are some other businesses where accounting is critical? Well, farmers need to follow careful accounting procedures. Many of them run their farms year to year by taking loans to plant the crops. If it's a good year, a profitable one, then they can pay off their loan; if not, they might have to carry the

[Stan Mathews]

Re: Procedure for establishing db of text files.

For a table "txts" with fields "field1","field2","field3","field4", and "field5" and the text files in c:\temp

txt_list = filefind.get("c:\temp\*.txt",FILE_FIND_NORMAL,"PN")
tbl = table.open("txts",FILE_RW_EXCLUSIVE)
for each foo in txt_list
	contents = file.to_string(foo)
	struct_l = "Title:" $ contents
	if struct_l ' contents are structured
	tbl.enter_begin()
	tbl.field1 = strtran(extract_string(contents,"Title:","Word Count:"),crlf(),"")'title field
	tbl.field2 = val(extract_string(contents,"Word Count:","Summary:"))'word count field
	tbl.field3 = strtran(extract_string(contents,"Summary:","Keywords:"),crlf(),"")'summary field
	tbl.field4 = strtran(extract_string(contents,"Keywords:","Article Body:"),crlf(),"")'keywords field
	tbl.field5 = words(contents,word_number_get(contents,"Body:",crlf())+1,30000,.F.,crlf())'article body memo field
	tbl.enter_end(.T.)
	else
	tbl.enter_begin()
	tbl.field1 = word(contents,1,crlf())'title field
	tbl.field5 = alltrim(words(contents,2,30000,.F.,crlf()))'article body memo field
	tbl.enter_end(.T.)
	end if
next
tbl.close()

If you might have more than 30,000 lines in the Article Body, adjust the alltrim(words(contents,2,30000,.F.,crlf()))