OpenEuphoria

1. Storage of Euphoria variables

• Posted by Vinoba Apr 28, 2011
• 1723 views

If I want to use assembler, or Peek and Poke, is there information available on

1. How an atom is stored
2. How a sequence is stored
3. how an object is stored
4. How variable names are stored.

3. Re: Storage of Euphoria variables

• Posted by mattlewis (admin) Apr 28, 2011
• 1704 views

Take a look at the Euphoria Internals section of the manual. It documents a lot (not sure how complete it is, exactly) of what you're looking for. You can also look at the C headers that define the structs and macros.

• source/object.h:
  • struct s1
  • struct d

• source/execute.h
  • Macros for inspecting and manipulating objects

Note that the 64-bit versions introduced in 4.1 will be slightly different. Of course, if you have more specific questions, please ask...

Matt

5. Re: Storage of Euphoria variables

• Posted by zebra Apr 28, 2011
• 1709 views

Vinoba: 
 
This should get you started. 
It is not a complete program but it will give you an idea of 
how to do your conversion 
  

-- Using Euphoria Version 4 
include std/sequence.e 
include std/convert.e 
					 -- not open file in binary mode. 
integer fn = open("SomeFile.dbf", "rb")  -- some_file with the correct extent 
-- 
if fn = -1 then 
puts(1,"can't open file") 
end if 
-- 
sequence header = {}, field_descrp_array = {}, record = {} 
 
-- BYTE....CONTENTS........MEANING .................................DETAILS 
-- 0.......1 byte ........ dBASE III version number................(03H without a .DBT file) 
-- ................................................................(83H with a .DBT file) 
-- 1-3.....3 BYTES ........date of last update ....................(YY MM DD) in binary format 
-- 4-7.....32 bit number...number of records in data file 
-- 8-9.....16 bit # .......length of header structure 
-- 10-11...16 bit #........length of the record 
-- 12-31...20 bytes........reserved bytes (version 1.00) 
-- 
-- read 32 the above header bytes 
-- 
for i = 0 to 31 do 
header &= getc(fn) 
end for 
-- Now the sequence holds the header has been read into 
-- a sequence NOTE THE SEQENCE STARTS AT ONE NOT ZERO 
-- 
-- header[1]      contains version number 
-- pad with zeros because bytes_to_int converts four bytes 
atom version = bytes_to_int({0,0,0}&header[1]) -- note 1 byte with 3 zeros 
-- header[2]      contains YY 
atom year = bytes_to_int({0,0,0}&header[2]) 
-- header[3]      contains MM 
atom month = bytes_to_int({0,0,0}&header[3]) 
-- header[4]      contains DD 
atom day = bytes_to_int({0,0,0}&header[4]) 
-- header[5..8]   contains number of records 
atom no_rec = bytes_to_int({0,0}&header[5..8]) 
-- header[9..10]  contains length of header structure 
atom len_hdr = bytes_to_int({0,0}&header[9..10]) 
-- header[11..12] contains length of record 
atom len_rec = bytes_to_int({0,0}&header[11..12]) 
-- header[13..33] contains reseved bytes 
-- *** reseved bytes are not used 
-- 
-- Now we read the array of field descriptor. 
-- NOTE THAT THE FILE IS STILL OPEN 
-- SO WE WILL START ON THE VERY NEXT BYTE. 
-- 
-- 32-n....32 bytes each...field descriptor array..................See Table 2 
-- n+1.....1 byte..........field terminator........................0D (HEX) CR, no LF 
-- 
atom byte = -1 
while 1 do 
  for r = 0 to 31 do 
    byte = getc(fn) 
    if byte = -1 then exit end if -- end of file 
    field_dscrip &= getc(fn) 
  end for 
  if byte = -1 then exit end if -- end of file 
  field_descrp_array &= {field_dscrip} 
end while 
-- Close the file IF YOUR done 
close(fn) 
 
 
-- The array will look like this. 
-- field_descrp_array { {field_dscrip#1},{field_dscrip#2},{field_dscrip#3} } 
-- You would go through and  decode each field_dscrip like this: 
--- 
 
 
-- Table 2: A FIELD DESCRIPTOR: 
-- BYTE CONTENTS MEANING DETAILS 
-- 0-10 11 bytes field name in ASCII zero-filled 
-- 11 1 byte field type in ASCII C N L D or M 
-- 12-15 32 bit # field data address address is set in memory 
-- 16 1 byte field length binary 
-- 17 1 byte field decimal count binary 
-- 18-31 14 bytes reserved bytes (version 1.00) 
 
-- field_dscrip[1..11]  contains name 
sequence name = field_dscrip[1..11] 
for i = 11 to 1 do -- trim off the zeros 
if name[i] = 0 then name = name[1..$-1] else exit end if 
end for 
-- field_dscrip[13..16] contains field data address 
atom  address = bytes_to_int(field_dscrip[13..16]) 
-- field_dscrip[17]     contains field length binary 
atom field_len = bits_to_int(field_dscrip[17]) 
-- field_dscrip[18]     contains decimal count binary 
atom dec_cnt = bits_to_int(field_dscrip[18]) 
-- field_dscrip[19..32] contains 14 bytes reserved bytes 
-- ***** NOT USED 

7. Re: Storage of Euphoria variables

• Posted by DerekParnell (admin) Apr 28, 2011
• 1735 views

Are you trying to access these data items when they are in RAM or when they are on disk?

If RAM, then you do not need the exact implementation details that Euphoria uses - just use standard Euphoria access methods. Using peek/poke is not advised because it would be generally slower than the optimized Euphoria syntax, and the internal implementation could change without notice in future Euphoria versions.

It is not possible to get the RAM address of the stored data from your Euphoria code so even if you knew how it was stored, you still couldn't get access to it.

If you want to use peek() to access your data, you will also need to use poke() to store it and manage it yourself. You will not be able to use in-line assembler to access native Euphoria data.