1 | initial version |

Sage has some weaknesses when it comes to dealing with assumptions.

The best I can think of is the following:

```
sage: var('n eps')
sage: f(n) = 1 /(n + 7)
sage: assume('eps > 0')
sage: N = (1 / eps) - 7
sage: assume(n > N)
sage: sol = solve([abs(f(n)) < eps],n)
sage: sol
[[n < -7, -eps*n - 7*eps - 1 > 0],
[n == -7, -1 > 0],
[-7 < n, eps*n + 7*eps - 1 > 0]]
```

From there you can continue manipulating each of the "solutions".

Let's look at the second one. It's a bit disappointing that `solve`

did not suppress it, given that `-1 > 0`

has no solution. But if you solve it again, you get an empty list, indicating "no solution" (beware that it might also mean "no solution found").

```
sage: a, b, c = sol
sage: b
[n == -7, -1 > 0]
sage: solve(b,n)
[]
```

For `a`

, it's also disappointing that it doesn't get suppressed, since it's easy to derive that `n > -7`

.

For `c`

, it's also disappointing that it doesn't get simplified, since the second condition follows from the first, and could therefore be suppressed.

Copyright Sage, 2010. Some rights reserved under creative commons license. Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.