integer constraints in lp format do not seem to be respected

[Dhakir]

Inequality constraints and integer variables give clearly incorrect results, which are not explicitly mentioned in the documentation.

For instance, trying to solve the following program in lp format:

min: x;

x > 0;

int x;

Results in:

Value of objective function: 0

Actual values of the variables:
x 0

Which is clearly incorrect.
The solution, which involves using some artificial boundaries such as "x > 0.00000001" instead of "x > 0", is NOT intuitive and nowhere mentioned in the documentation. This is extremely frustrating for novice users.

If there are limitations which prevent lp_solve from correctly parsing the given input program without having to add some residual value to "0", then at least it should be mentioned in the documentation.

[Martin Pokorny (martin-truffulatree)]

This is actually described in the lp_solve documentation under the description of the LP file format, where it states:

<constraint> is an optional constraint name followed by a colon plus
  a linear combination of variables and constants or (just one)
  constraint name followed by a colon (a range) or (just one) variable
  name without a colon (a bound), followed by a relational operator,
  followed again by a linear combination of variables and constants,
  ending with a semicolon. The relational operator can be any of the
  following: "<" "<=" "=" ">" ">=". There is no semantic difference
  between "<" and "<=" nor between ">" and ">=" (even for integer
  variables!).